近五年移动机器人轨迹规划的国内外研究现状

No abstract available

In order to better meet the practical application needs of mobile robots, this study innovatively designs an autonomous obstacle avoidance and trajectory planning control strategy with low computational complexity, high cost-effectiveness, and the ability to quickly plan a collision-free smooth trajectory curve. This article constructs the kinematic model of the mobile robot, designs a dual-loop trajectory tracking control strategy for position control law and attitude control law algorithms, and improves the traditional artificial potential field method to achieve a good obstacle avoidance strategy for mobile robots. Based on the dual-loop trajectory tracking control and the improved artificial potential field method, the autonomous obstacle avoidance and trajectory planning scheme of the mobile robot is designed, and closed-loop stability verification and analysis are conducted on the overall control system. And through the detailed simulation and experiments, the advantages of the proposed method in trajectory tracking accuracy and motion stability compared to the existing methods are verified, showing good effectiveness and feasibility and laying a good foundation for the application of mobile robots in practical complex scenes.

ABSTRACT

Trajectory planning is important for ground robots to achieve safe and efficient autonomous navigation in unstructured off-road environments. Most existing methods treat each terrain as a single type. However, in the real world, a ground usually consists of hybrid terrains. In this work, we propose a novel trajectory planning network that handles hybrid terrain. To further enhance safety, we have designed a self-correcting structure based on historical planning data. This structure can correct the trajectory when an inappropriate one is planned. To train the network, we introduce a two-stage training scheme based on Offline Meta-Reinforcement Learning, which can train the network with pre-collected non-optimal datasets and reduce the occurrence of hazardous planning. The proposed approach has been evaluated on both simulated datasets and a real robot platform. Compared to state-of-the-art baseline methods, the proposed approach reduces hazardous planning by 59.3% in hybrid terrains.

This article is concerned with the problem of planning optimal maneuver trajectories and guiding the mobile robot toward target positions in uncertain environments for exploration purposes. A hierarchical deep learning-based control framework is proposed which consists of an upper level motion planning layer and a lower level waypoint tracking layer. In the motion planning phase, a recurrent deep neural network (RDNN)-based algorithm is adopted to predict the optimal maneuver profiles for the mobile robot. This approach is built upon a recently proposed idea of using deep neural networks (DNNs) to approximate the optimal motion trajectories, which has been validated that a fast approximation performance can be achieved. To further enhance the network prediction performance, a recurrent network model capable of fully exploiting the inherent relationship between preoptimized system state and control pairs is advocated. In the lower level, a deep reinforcement learning (DRL)-based collision-free control algorithm is established to achieve the waypoint tracking task in an uncertain environment (e.g., the existence of unexpected obstacles). Since this approach allows the control policy to directly learn from human demonstration data, the time required by the training process can be significantly reduced. Moreover, a noisy prioritized experience replay (PER) algorithm is proposed to improve the exploring rate of control policy. The effectiveness of applying the proposed deep learning-based control is validated by executing a number of simulation and experimental case studies. The simulation result shows that the proposed DRL method outperforms the vanilla PER algorithm in terms of training speed. Experimental videos are also uploaded, and the corresponding results confirm that the proposed strategy is able to fulfill the autonomous exploration mission with improved motion planning performance, enhanced collision avoidance ability, and less training time.

A velocity obstacle crossing the paths of Mobile robots is a general situation in both indoor and outdoor environments. Velocity obstacles reduce or alter the speed of the robot. The prediction of exact trajectory for the robot motion by the currently available methods is to be improved a lot. This gap is to be addressed. This manuscript proposes two multi-objective optimization methods based on evolutionary algorithms to bridge the gap. Evolutionary algorithms - Multi-objective Differential Evolution (MODE) and Elitist non-dominated sorting Genetic Algorithm (NSGA-II) are proposed in this paper. A car like mobile robot is designed, fabricated and experimented. Travel time and energy spent are considered as performance criteria in the problem solved by the proposed algorithms. Dynamic parameters' limitations, geometrical parameters' constraints and kinematic characteristics' bounds are considered in the problem. Both fixed and moving obstacles were present near the robot in its environment. So, motion of both obstacles was considered by the algorithms while planning the path of the robot. A third order NURBS function is used to characterize the mobile robot's trajectory. The optimal trajectory for the mobile robot was found by both NSGA-II and MODE. Experiments also conducted to validate the algorithms' results. Experimental validation proves that both NSGA-II and MODE are good algorithms to consider for optimal trajectory planning of mobile robots while considering stationary obstacles.

The field of autonomous mobile robots has been gaining significant attention in various industries and research domains. As the future of robotic process automation unfolds, there is an increasing demand for precise robot movement in terms of collision avoidance and trajectory planning. This paper presents a camera-based autonomous mobile robot system that addresses these requirements. The proposed system utilizes a deep learning variational autoencoder with a spatio-temporal model for image analysis processing. This approach enables the system to effectively analyze and understand the visual information. By leveraging deep learning techniques, the system can extract meaningful features and representations from the images, facilitating accurate perception and understanding of the robot's surroundings. This paper contributes to the advancement of autonomous mobile robot systems by proposing a deep learning techniques with reinforcement learning algorithms. The approach offers promising possibilities for enhancing the control and interaction capabilities of mobile robots in real-world scenarios.

This paper proposes an integrated framework of trajectory planning and control for autonomous mobile robots (AMRs) in an intra-logistic scenario, where humans and robots share the same indoor workspace. The proposed framework consists of perception and trajectory planning. To perceive human motion, information from the RGB-D camera is used to detect human positions. Then a model-based tracking approach is employed to track multiple people’s motion and estimate their speed over a short time horizon. For trajectory planning, a Model Predictive Control (MPC) based scheme is adopted to generate motor commands while taking into account energy efficiency, safety, and addressing human-aware proximity constraints. To convexify the nonlinear collision avoidance constraints, Sequential Convexification Programming is employed. The simulation and experimental results demonstrate that the proposed method can be implemented in real-time and efficiently avoid people in advance.

Recent developments in the field of robotics have led to discussions surrounding the human-robot coexistence environments including homes and modern factories. Focusing on the application of mobile robots, the focus of this research is to enhance their performance in dynamic scenarios. To effectively plan the robot’s path to avoid pedestrians, a machine learning algorithm is employed to predict the future trajectory of pedestrians, thus improving the accuracy of forecasting their multi-modal motion. The existing prediction methods primarily rely on pedestrian history and current movement attributes to predict future movement, they often overlook the impact of static obstacles on pedestrian movement decision. Therefore, in this study, a static obstacles probability description generative adversarial network (SOPD-GAN) is proposed. The static obstacles probability description (SOPD) represents the future movement space of pedestrians and assesses the likelihood of entry. Additionally, we incorporate pedestrian historical trajectory information using LSTM, and combine it with SOPD to form the generator model. The training of this model is carried out using a generative adversarial network (GAN), which is referred to as SOPD-GAN. In addition, we also introduce an improved dynamic window approach (IDWA) for robot path planning in dynamic scenarios based on pedestrian trajectory prediction. In order to validate the efficacy of our approach, we conduct experiments in real scenarios and compare the model with existing baselines. The results show that this method can construct a suitable prediction model with high accuracy. Specifically, our method achieved an accuracy of 0.0881 and 0.0691 in FDE and AEE of predicting pedestrian trajectory, surpassing the baseline method by 20% and 14%.

: In this paper, our goal is to present the autonomous cone placing robot developed at SzéchenyiIstván University (Gy ő r, Hungary) and the main steps and parts of its design and preparation. Within this, rather complex task-sequence, we discuss the logic of software operation (embedded in the ROS 2 ecosystem), the main issues of environmental representation and we focus especially on the trajectory planner part of the entire system. The implemented algorithms-including our own innovative ideas are Dijkstra, A *, Hybrid A *, DWA and Elastic band.

The task of local trajectory planning for an autonomous wheeled robotic platform in cluttered indoor environment is considered. Such environment might include narrow passages, which width is less than the length of the platform. Therefore, it is not possible to apply standard approach, when the obstacles are inflated with the maximum radius of the platform. We propose a novel approach based on numerical solution of nonlinear model predictive control task. Oblong shape of the platform is approximated with a high-order ellipse. We define differentiable sigmoid po-tential function, which may be computed for any point of the workspace given position and orientation of the plat-form. This function is small far from the platform, and very high inside the platform; it increases when moving to-wards the robot. The value of this potential function is computed for the set of the support obstacle points and add-ed to the cost function. This function serve as a penalty for collision with obstacles or coming too close to them. We develop an algorithm for the mapping support points onto occupancy grid, which provide collision avoidance. We apply Acados open library, which implement numerical solution of nonlinear model predictive task with sequential quadratic programming. Our approach is implemented as a local planner for the collaborative mobile platform. The experiments were made in artificial maze and in real office environment with narrow passages. Proposed ap-proached allowed the robot to come through the passages that were 10-20 cm wider than the platform. Computa-tion time was around 20 milliseconds.

No abstract available

Abstract Robotics with artificial intelligence techniques have been the center of attraction among researchers as it is well equipped in the area of human intervention. Here, the krill herd (KH) optimization algorithm is modified and hybridized with a fuzzy logic controller to frame an intelligent controller for optimal trajectory planning and control of mobile robots in obscure environments. The controller is demonstrated for single and multiple robot’s trajectory planning. A Petri-net controller has also been added to avoid conflict situations in multi-robot navigation. MATLAB and V-REP software are used to simulate the work, backed with real-time experiments under laboratory conditions. The robots efficiently achieved the goals by tracing an optimal path without any collision. Trajectory length and time spent during navigation are recorded, and a good agreement between the results is observed. The proposed technique is compared against existing research techniques, and an improvement of 14.26% is noted in terms of path length.

Trajectory optimization is the series of actions that are taken into consideration in order to produce the best path such that it improves the overall performances of physical properties or reduces the consumption of the resources where the restriction system remains maintained. In this paper, first, a compact mathematical model for a symmetrical annular-shaped omnidirectional wheeled mobile robot (SAOWMR) is derived and verified. This general mathematical model provides an opportunity to conduct research, experiments, and comparisons on these omnidirectional mobile robots that have two, three, four, six, or even more omnidirectional wheels without the need to switch models or derive a new model. Then, a new computationally efficient method is proposed to achieve improvements in the trajectory planning optimization (TPO) for a SAOWMR. Moreover, the proposed method has been tested in collision-free navigation by incorporation of the path constraints. Numerical tests and simulations are presented aiming to ensure the efficiency and effectiveness of the proposed method.

This paper proposes a novel trajectory planning approach based on time elastic band to solve the problem of dynamic obstacle avoidance of mobile robot. Uncertain factors in the scenario need to be considered in trajectory planning. Thus, this approach includes multiple constraints, such as robot motion speed, motion state, and obstacles. First, to solve the optimal speed of the mobile robot, the workspace potential field must be established, and environmental information should be obtained to constrain the robot speed. Second, a costmap needs to be established to detect dynamic obstacles, and obstacle avoidance strategies based on the relative motion relationship between dynamic obstacles and the robot should be proposed to realize dynamic obstacle avoidance. Finally, by combining multiple constraints, the collision-free trajectory planning from the start point to the target point is completed, and the mobile robot realizes collision-free smooth motion. Experimental results show that this approach has satisfactory obstacle avoidance planning effects and superior kinematics characteristics and improves the comfort and safety of the mobile robot.

No abstract available

Trajectory planning enables Autonomous Mobile Robot (AMR) to have intelligence and avoid a collision in the interaction with obstacles. However, in scenes with multiple obstacles, most of the existing methods cannot minimize the collision risk. It is because that these methods do not distinguish the importance of the obstacles in the scene. Therefore, in this paper, we proposed an Obstacle-Centered Trajectory Planning (OCTP) method to solve the problem. In our method, a novel collision risk evaluation model is constructed, which considers the importance of each obstacle. In addition, a sliding-window-based key points interpolation method is used to smooth the velocity profile obeying constraints of collision risk and curvature. Finally, a comparison with the baseline method is performed. The experimental results show that the proposed method can effectively reduce AMR's collision risk in interacting with obstacles.

No abstract available

Compared to four-wheel independent drive and steering (4WID4WIS) mobile robots, three-wheel independent drive and steering (3WID3WIS) mobile robots are more cost-effective due to their lower cost, lighter weight, and better handling performance, even though their acceleration performance is reduced. This paper proposes a dynamic model predictive control (DMPC) controller for trajectory tracking of 3WID3WIS mobile robots to simplify the computational complexity and improve the accuracy of traditional model predictive control (MPC). The A* algorithm with a non-point mass model is used for path planning, enabling the robot to navigate quickly in narrow and constrained environments. Firstly, the kinematic model of the 3WID3WIS mobile robot is established. Then, based on this model, a DMPC trajectory tracking controller with dynamic effects is developed. By replacing MPC with DMPC, the computational complexity of MPC is reduced. During each control period, the prediction horizon is dynamically adjusted based on changes in trajectory curvature, establishing a functional relationship between trajectory curvature and prediction horizon. Subsequently, a comparative study between the proposed controller and the traditional MPC controller is conducted. Finally, the new controller is applied to address the trajectory tracking problem of the 3WID3WIS mobile robot. The experimental results show that DMPC improves the lateral trajectory tracking accuracy by 62.94% and the heading angle tracking accuracy by 34.81% compared to MPC.

Global path planning is a critical technology in the field of autonomous mobile robot navigation. Serving as the upper-layer component of path planning, it provides the global reference path for the local trajectory planning module. However, the majority of conventional methods focus solely on optimizing path length, which can lead to redundant obstacle avoidance maneuvers by the lower-layer local planner or even planning failure. Furthermore, graph-searching methods commonly suffer from prolonged computation times and low efficiency. To address these challenges, this paper proposed a global path planning method based on deep reinforcement learning that integrates artificial potential fields. The method expanded the network structure of Soft Actor-Critic (SAC) by employing the constructed potential field to conduct supervised learning on two additional critic networks. Subsequently, the predicted values from the critic network were integrated into the actor network to guide agents in choosing states with smaller potential field values. Additionally, to mitigate the time cost of retraining due to changes in the global environment, a risk assessment module employing Monte Carlo random sampling was incorporated. The computed risk value was subsequently integrated into the network as the new state. Experimental results show that our method reduces computation time by 38.64% compared to conventional methods. The convergence is 40.48% faster and the path potential value is 95.72% lower than other reinforcement learning methods.

In this paper, a real-time obstacle avoidance approach and a trajectory planning method are proposed to avoid collisions to move an object jointly by a human and an omnidirectional mobile robot. Different from many existing approaches of local trajectory planning, the proposed method called Multiscale Local Perception Region Approach (MLPRA) is specially designed for obstacle avoidance of omnidirectional wheeled robots with direction constraints of human guidance, which can respond fast to dynamic obstacles and guarantee the safety of the robot. To solve the problem that the position of the human hand is difficult to obtain by visual sensing due to visual occlusion, a simple mechanism that can indirectly measure the change of the position of the human hand is designed, and further a method to follow the human’s intention based on this mechanism is proposed. Finally, the simulation environment on the Gazebo simulation platform is built to verify the feasibility and effectiveness of our proposed methods. Experimental results show that after embedding proposed methods into the omnidirectional mobile robot, obstacles can be effectively avoided in co-transporting processes.Note to Practitioners—The motivation of this paper is focusing on obstacle avoidance of omnidirectional mobile robots in human-robot co-transporting, application scenarios concentrated in factories and logistics warehouses, such as the mobile robot collaborates with a human (a robot teleoperated by human) transporting a table. The existing obstacle avoidance algorithm can be regarded as a local trajectory planning problem, which cannot directly adapt to co-transporting tasks in real-time. Considering the specific task if the robot cannot cooperate with human in real-time, the carried object will fall. In this paper, a local trajectory planning method is proposed, regarding the obstacle generating virtual repulsive force, regarding the human action on the robot as the virtual gravitational force, and the mobile robot under two virtual forces not only follows human action in co-transporting, but also avoids obstacles. The method shows quick calculation and good real-time performance. The method proposed is evaluated by co-transporting in Gazebo and experiments based on omnidirectional mobile robots, and ensured that the object being carried does not fall and robot can avoid collisions in the unknown environments without positioning systems.

This paper focussed on the development of a dynamic and efficient obstacle avoidance path planning algorithm based on ORCA-DWA algorithm, which combines the Optimal Reciprocal Collision Avoidance (ORCA) algorithm and an improved Dynamic Window Approach (DWA), for improving the quality and efficiency of globally planned paths and ensuring obstacle avoidance between robots for local path planning. This combined ORCA-DWA approach effectively combines the speed of DWA planning with the preferred speed of the ORCA algorithm, it not only solves the problem of the ORCA algorithm's difficulty in determining the preferred speed, but also does not deviate from its optimal trajectory while avoiding obstacles. Additionally, an improved dynamic windowing method is proposed to enhance the adaptability to the environment. As a result, the mobile robot can not only use the DWA algorithm to achieve global path optimisation during navigation, but also achieve obstacle avoidance with the shortest time and path while following the robot's own constraints and considering the robot's radius. Simulation results prove that the method can greatly reduce the length and time of path planning and show that this new algorithm can make the robot's speed smoother.

In mobile robot path planning, the traditional A* algorithm suffers from high path redundancy and poor smoothness, while the Dynamic Window Approach (DWA) tends to deviate from the global optimal path and has low efficiency in avoiding dynamic obstacles when integrated with global path planning. To address these issues, a smoothing optimised A*-guided DWA fusion algorithm (SOA-DWA) is proposed in this paper. Firstly, the A* algorithm was improved by introducing a path smoothing strategy and path pruning mechanism, generating a globally optimal path that complied with the vehicle kinematic constraints. Secondly, three sub-functions were introduced into the evaluation function of the DWA algorithm: the distance evaluation between the reference trajectory and the global path, the path direction evaluation, and the dynamic obstacle avoidance evaluation, to enhance the real-time performance of dynamic obstacle avoidance and the consistency of the global path. The SOA-DWA algorithm ensured that the mobile robot could effectively avoid obstacles in complex environments without deviating from the global optimal path. Thirdly, experimental results show that in a static environment, the path length and turning angle of the SOA-DWA algorithm are reduced by an average of 13.3% and 16.25%, respectively, compared with the traditional algorithm. In a dynamic environment, the path length and turning angle are reduced by an average of 10.5% and 14.5% compared to the traditional DWA algorithm, respectively, significantly improving the smoothness of the path and driving safety. Compared to the existing fusion algorithm, the SOA-DWA algorithm reduces the path length by an average of 10.1%, improves planning efficiency by an average of 42%, and effectively enhances obstacle avoidance efficiency. Finally, the effectiveness of the improved algorithm proposed in this paper was further verified by mobile robot experiments.

Abstract The transformation of existing production and logistics halls into flexibly configurable factories requires the ability of autonomous mobile robots to navigate efficiently in an unstructured and changing environment. Conflicting situations between several robots due to collision avoidance when moving close to each other can be resolved via the central control system or in a distributed manner. In this paper, we address distributed coordinated trajectory planning of mobile robots with communication to efficiently resolve collision avoidance scenarios without a central control unit. We simulatively investigate coordinated planners that solve optimal control problems with a sliding time horizon. Further, the planners differ in whether a centralized optimal control problem in the coupled state space of the robots or multiple decentralized optimization problems in the state space of one robot are formulated. From another point of view, the methods vary in a cooperative or non-cooperative property which manifests itself in the consideration or neglection of the other vehicles’ intentions within the cost function. Further, we not only consider homogeneous mobile robots that are modeled by the same vehicle model and the same parameterization but also heterogeneous robots with different physical limitations and dimensions. We present simulation results of symmetric and non-symmetric collision avoidance scenarios with 2–8 mobile robots. As a comparison criterion, we use the average duration until all vehicles have reached their destinations. The results show that the cooperative planners resolve homogeneous scenarios and the non-cooperative planners resolve heterogeneous ones involving vehicles having different dimensions with a lower average scenario duration.

To address the issues of slow convergence speed and poor path quality of the traditional Rapidly-exploring Random Tree Star (RRT*) algorithm in complex environments, this paper proposes a Multi Strategy Bidirectional RRT* (MS-BI-RRT*) algorithm for efficient mobile robot path planning. In the new node generation phase, an expansion mode scheduling mechanism based on dynamic goal bias probability and expansion feedback is designed to enable adaptive switching among multiple expansion modes, thereby improving expansion efficiency. Meanwhile, a dynamic step size adjustment method based on local obstacle density is introduced to enhance expansion stability. During the parent node rewiring phase, a multi-factor path cost function is constructed to optimize parent node selection, thereby improving path quality. In the post-processing phase, a Bézier curve-based smoothing strategy is employed to improve trajectory continuity and dynamic controllability. Simulation results in five typical environments show that, compared with RRT*, BI-RRT*, APF-RRT*, BI-APF-RRT*, and GB-RRT*, MS-BI-RRT* algorithm reduces the average execution time by 77.50%, decreases the number of nodes by 76.41%, shortens the path length by 4.37%, and achieves a 100% success rate in all environments. These results demonstrate that the proposed method significantly improves convergence speed, path quality, and environmental adaptability, while exhibiting superior robustness.

Real-time robotic teleoperation is critical for complex environment operation and imitation learning, yet existing methods face limitations in user-friendliness, robot performance adaptation, and multi-modal scalability. This paper proposes a mobile phone teleoperation system that leverages smartphone augmented reality (AR) for real-time 6-degree-of-freedom (DoF) pose tracking and ensures safe and smooth motion through trajectory planning based on quadratic programming (QP). A configurable mobile app interface enhances scalability by supporting auxiliary joystick and button controls. Experimental validation on a quadruped mobile manipulator demonstrates sub-20 mm AR pose tracking accuracy across environments, end-to-end latency below 100 ms, and smooth task execution via QP-based planning, including object gripping and door handle pulling. The results demonstrate the potential of the framework for cost-effective, user-friendly, and extensible teleoperation in dynamic mobile manipulation scenarios.

: At present, energy consumption is one of the main bottlenecks in autonomous mobile robot development. To address the challenge of high energy consumption in path planning for autonomous mobile robots navigating unknown and complex environments, this paper proposes an Attention-Enhanced Dueling Deep Q-Network (AD-Dueling DQN), which integrates a multi-head attention mechanism and a prioritized experience replay strategy into a Dueling-DQN reinforcement learning framework. A multi-objective reward function, centered on energy efficiency, is designed to comprehensively consider path length, terrain slope, motion smoothness, and obstacle avoidance, enabling optimal low-energy trajectory generation in 3D space from the source. The incorporation of a multi-head attention mechanism allows the model to dynamically focus on energy-critical state features—such as slope gradients and obstacle density—thereby significantly improving its ability to recognize and avoid energy-intensive paths. Additionally, the prioritized experience replay mechanism accelerates learning from key decision-making experiences, suppressing inefficient exploration and guiding the policy toward low-energy solutions more rapidly. The effectiveness of the proposed path planning algorithm is validated through simulation experiments conducted in multiple off-road scenarios. Results demonstrate that AD-Dueling DQN consistently achieves the lowest average energy consumption across all tested environments. Moreover, the proposed method exhibits faster convergence and greater training stability compared to baseline algorithms, highlighting its global optimization capability under energy-aware objectives in complex terrains. This study offers an efficient and scalable intelligent control strategy for the development of energy-conscious autonomous navigation systems.

Planning in mobile robots is an essential task because it helps to achieve efficient movement in terms of time, computational resources, and safety. Warehouse robots, for example, are equipped with sensors and cameras to avoid obstacles and to move products around the warehouse pick and pack orders. In this project, we have used path-planning algorithms based on Reinforcement Learning (RL), which include Q-learning (QL), State-Action-Reward-State-Action (SARSA), and Expected SARSA (ESARSA). We evaluated these algorithms on a benchmark dataset with different sizes and obstacle densities. Our findings show that QL produces a more optimal path, but the path is risky since it is close to obstacles. On the other hand, SARSA produces a safer path. However, in terms of convergence speed, SARSA is slow, while ESARSA is faster and more stable but with extra computations.

Mobile robot motion planners rely on theoretical models to predict how the robot will move through the world. However, when deployed on a physical robot, these models are subject to errors due to real-world physics and uncertainty in how the lower-level controller follows the planned trajectory. In this work, we address this problem by presenting three methods of incorporating stochastic controller behavior into the recombinant search space of the Kinodynamic Efficiently Adaptive State Lattice (KEASL) planner. To demonstrate this work, we analyze the results of experiments performed on a Clearpath Robotics Warthog Unmanned Ground Vehicle (UGV) in an off-road, unstructured environment using two different perception algorithms, and performed an ablation study using a full spectrum of simulated environment map complexities. Analysis of the data found that incorporating stochastic controller sampling into KEASL leads to more conservative trajectories that decrease predicted collision likelihood when compared to KEASL without sampling. When compared to baseline planning with expanded obstacle footprints, the predicted likelihood of collisions becomes more comparable, but reduces the planning success rate for baseline search.

The main objective of the Mobile Robot Path Planning Problem is to find the optimal waypoints for a mobile robot with obstacles collision-free. This is a very complicated and needed task in robotic. Basically, planning rapidly the optimal task will increase the performance of the robot by increasing the speed to reach the target position and reducing energy conception. In this research work, the innovative technique namely Dhouib-Matrix-SPP (DM-SPP) is studied with eight movement directions as well as four. DM-SPP is a very rapid method built on the contingency matrix navigation and needing only n iterations to create the optimal path (where n is the number of nodes). The simulation results on several complicated case studies (varying from (20 x 20) grid map to (80 x 80) grid map) prove that DM-SPP can rapidly create an accurate trajectory with obstacles collision-free. Moreover, the proposed technique is compared with the very recently designed artificial intelligence approaches. The results of this comparison proved that the novel DM-SPP is the fastest approach: For example, it is (289.325) times rapider than the A* algorithm, (156.769) times faster than the Improved A* method, (127.901) times speedier than the Bidirectional A* technique, (69.586) times quicker than the Improved Bidirectional A* algorithm and (45.671) times rapider than the Variable Neighborhood Search BA* metaheuristic. These findings underline the speed of the proposed DM-SPP optimization technique and emerge the applicability of DM-SPP as a reliable option for the trajectory optimization.

To enhance the overall performance of autonomous mobile robot in path planning and tracking control, including planning efficiency, path quality, and obstacle avoidance capability, this paper introduces an innovative approach to path planning and tracking control by combining an enhanced A* algorithm with Model Predictive Control (MPC). Primarily, a dynamic weighting factor is incorporated into the conventional A* algorithm to enable adaptive modulation of the cost function. Concurrently, a mechanism for repeated node detection and updating within the Open list is devised to ensure path consistency and effectively reduce redundant node expansions. Subsequently, the initial path is refined using B-spline curve smoothing, thereby enhancing its continuity and feasibility. In the final stage, during path tracking, the MPC algorithm is employed for trajectory control, with Control Barrier Function (CBF) embedded into the control model to facilitate effective obstacle avoidance. Simulation results demonstrate that, in terms of path planning, the proposed improved A* algorithm significantly enhances path generation efficiency and quality, producing shorter, smoother paths with fewer redundant nodes. In terms of path tracking, the incorporation of CBF into the MPC controller improves obstacle avoidance capability against dynamic obstacles while maintaining tracking accuracy. Overall, the proposed method outperforms the traditional A* algorithm and MPC combination in both the path planning and tracking control stages.

This paper proposes a mobile robot motion control and planning system for trajectory tracking and obstacle avoidance in a prior unknown robot environment. The proposed system has two-level control and planning architecture: the higher is used to generate a path, while the lower provides the control actions that drive the robot. The planning level represents a reactive planer which determines on-line way-points during the robot’s movement towards the target and allowing the robot to move autonomously through an environment without colliding with obstacles. The main objective of this algorithm is to reduce the number of obstacles that are taken into consideration when determining the intermediate target point (way-points) in the movement towards the target location. This proposed algorithm is based on the concept of calculating the intersection of the variable target circle and the robot perception circle (VTPC), as well as attractive and repulsive forces. The lower level includes a fuzzy logic controller that drives the robot along generated online trajectory. It compares the current position of the mobile robot with the desired position, generating the appropriate linear speeds for the robot’s wheels to reach the target point in the shortest possible time. A series of simulations demonstrate its effectiveness in generating and executing the paths in various unknown robot environments.

In crowded environments, such as downtown areas on weekends or office corridors during commute hours, it is challenging for an autonomous mobile robot to reach its destination when navigating among moving pedestrians. This challenge requires advanced techniques that enable robots to understand and interact with human behavior, leveraging a wide range of real-world data. We propose two key components to address this crowd navigation challenge. First, the behavior of pedestrians around the robot is modeled using data collected by the robot. Specifically, K-means++ is applied to the trajectory data to identify distinct pedestrian behavior patterns; Subsequently, a transformer-based VAE reproduces pedestrians’ reactive motions to robot actions. Second, a ResNet-18 + A3C-based deep reinforcement learning algorithm trains the robot to generate safe navigation actions in a simulation environment that reflects actual pedestrian behavior. We validate our approach through real-world experiments conducted at a science museum, demonstrating significant improvements over existing methods.

Compact heavy-duty skid-steer robots are increasingly used for city logistics and intralogistics tasks where high payload capacity and stability are required. However, their limited maneuverability and non-negligible turning radius challenge conventional waypoint-tracking controllers that assume unconstrained motion. This paper proposes a curvature-constrained trajectory planning and control framework that guarantees geometrically feasible motion for such platforms. The controller integrates an explicit curvature limit into a finite-state machine, ensuring smooth heading transitions without in-place rotation. The overall architecture integrates GNSS-RTK and IMU localization, modular ROS 2 nodes for trajectory execution, and a supervisory interface developed in Foxglove Studio for intuitive mission planning. Field trials on a custom four-wheel-drive skid-steer platform demonstrate centimeter-scale waypoint accuracy on straight and curved trajectories, with stable curvature compliance across all tested scenarios. The proposed method achieves the smoothness required by most applications while maintaining the computational simplicity of geometric followers. Computational simplicity is reflected in the absence of online optimization or trajectory reparameterization; the controller executes a constant-time geometric update per cycle, independent of waypoint count. The results confirm that curvature-aware control enables reliable navigation of compact heavy-duty robots in semi-structured outdoor environments and provides a practical foundation for future extensions.

This paper presents a fusion algorithm based on the enhanced RRT* TEB algorithm. The enhanced RRT* algorithm is utilized for generating an optimal global path. Firstly, proposing an adaptive sampling function and extending node bias to accelerate global path generation and mitigate local optimality. Secondly, eliminating path redundancy to minimize path length. Thirdly, imposing constraints on the turning angle of the path to enhance path smoothness. Conducting kinematic modeling of the mobile robot and optimizing the TEB algorithm to align the trajectory with the mobile robot's kinematics. The integration of these two algorithms culminates in the development of a fusion algorithm. Simulation and experimental results demonstrate that, in contrast to the traditional RRT* algorithm, the enhanced RRT* algorithm achieves a 5.8% reduction in path length and a 62.5% decrease in the number of turning points. Utilizing the fusion algorithm for path planning, the mobile robot generates a superior, seamlessly smooth global path, adept at circumventing obstacles. Furthermore, the local trajectory meticulously conforms to the kinematic constraints of the mobile robot.

No abstract available

In modern warehouse automation, the platform environment to the automatic loading of goods in the container is basically through manual or large conveyor belt machinery to complete, the container is dark and narrow, easy to due to the goods scuffing or the target point is not clear resulting in reduced efficiency. The automatic loading of mobile robots can solve these problems, but at present there are fewer autonomous mobile robot solutions applied to the platform container environment. In this paper, A path planning algorithm is proposed for platform loading tasks, in which LiDAR scans the container to calculate the coordinates of the loading positions and obtains the best third-order Bezier curve based on particle swarm optimization. The experiment proves that the algorithm can effectively get the loading points in the container and fit an excellent smooth route to improve the efficiency of navigation.

Mobile robots are increasingly integral to diverse applications, with path-planning algorithms being essential for efficient and secure mobile robot navigation. Mobile robot path planning is defined as the design of the least time-consuming, shortest-distance, and most collision-free path from the starting point to the endpoint for the mobile robot’s autonomous movement. This study investigates and assesses two widely used algorithms in artificial intelligence (AI)—Improved Particle Swarm Optimization (IPSO) and Improved Genetic Algorithm (IGA)—for path planning of mobile robot navigation problems. In this work Manhattan movements are proposed as a distance formula to modify both algorithms in the path planning of the mobile robot navigation problem. Unlike the traditional GA and PSO, which can use horizontal search, the proposed algorithm relies on vertical search, which gives us an advantage. The results demonstrate the effectiveness of these modified algorithms in barrier detection and obstacle avoidance. Six different experiments were run using both improved algorithms to show their ability to achieve their goal and avoid obstacles in various scenarios with different complexities. Across various scenarios, the tested AI algorithms performed effectively, regardless of the map scale and complexity. This paper proposes a complete comparison between the two improved algorithms in different scenarios. The results show that the algorithms’ performance is influenced more by the density of walls and obstacles than by the size or complexity of the map.

In recent years, Automated Mobile Robots (AMRs) have gained significant attention in industry and research applications, requiring efficient path-planning algorithms to optimize task performance. While widely adopted, conventional Ant Colony Optimization (ACO) algorithms suffer from low convergence rates and delays in task execution, particularly in dynamic environments due to insufficient exploration of this context. However, traditional Ant Colony Optimization (ACO) algorithms, widely used for AMR path planning, exhibit limitations such as low convergence rates and redundant recalculations, particularly in environments with frequently changing obstacles. To address these challenges, this study proposes an Integrative Edge Cloud-Based Ant Colony Optimization (IECACO) algorithm. IECACO incorporates a novel path retrieval mechanism and edge cloud computing infrastructure to minimize redundant path computation and improve convergence efficiency. The proposed algorithm is tested within a simulated 2D occupancy grid environment using both a 4×4 map for controlled experiments and a 20×20 map for comparative evaluation against a prior Improved ACO (IACO) study. Experimental simulation results, based on 50 independent runs in settings, demonstrate that IECACO achieves at least 4.76% reduction compared to traditional ACO. Based on the observation of 10 independent runs between IECACO and IACO, IECACO leading a significant reduction in both static and dynamic settings. Although this study is conducted in a simulated environment, the findings lay a foundation for future real-world implementations.

Aiming at the scheduling problem of multi-autonomous mobile robots (AMR) in the box storage environment, the traditional dynamic programming (DP) algorithm has the disadvantage of low efficiency in solving the feasible path. To solve this problem, this paper establishes a reinforcement learning (RL) algorithm model with the goal of time optimization, which is used to improve the speed of path planning for multi-AMR simultaneous scheduling. In addition, combined with the advantages of the deep learning (DL) algorithm, the deep reinforcement learning (DRL) algorithm is used to effectively shorten the convergence time of the RL algorithm model training under high-dimensional and complex working conditions. The effectiveness of the DRL method is verified by comparing DP, RL, and DRL algorithm models in the simulation platform.

Mobile robots are intended to operate in a variety of environments, and they need to be able to navigate and travel around obstacles, such as objects and barriers. In order to guarantee that the robot will not come into contact with any obstacles or other objects during its movement, algorithms for path planning have been demonstrated. The basic goal while constructing a route is to find the fastest and smoothest route between the starting point and the destination. This article describes route planning using the improvised genetic algorithm with the Bezier Curve (GA-BZ). This study carried out two main experiments, each using a 20x20 random grid map model with varying percentages of obstacles (5%, 15%, and 30% in the first experiment, and 25% and 50% in the second). In the initial experiments, the population (PN), generation (GN), and mutation rate (MR) of genetic algorithms (GA) will be altered to the following values: (PN = 100, 125, 150, or 200; GN = 100, 125, 150; and MR = 0.1, 0.3, 0.5, 0.7) respectively. The goal is to evaluate the effectiveness of AMR in terms of travel distance (m), total time (s), and total cost (RM) in comparison to traditional GA and GA-BZ. The second experiment examined robot performance utilising GA, GA-BZ, Simulated Annealing (SA), A-Star (A*), and Dijkstra's Algorithms (DA) for path distance (m), time travel (s), and fare trip (RM). The simulation results are analysed, compared, and explained. In conclusion, the project is summarised.

Autonomous robots can be assigned with various tasks such as moving payload, analyzing terrain, and capturing data in an environment. For an Autonomous Mobile Robot (AMR) to execute such tasks the robot (Hussarion ROSbot) will require efficient algorithms and techniques to reference its current location. The robot is relative to surrounding obstacles in its predetermined path. The conducted research study explains the coordinated method used to successfully allow a robot to identify its position in the environment (Gazebo Simulation) and avoid obstructions with increasing velocity - contingent on nearby surroundings. The results show multiple robots individually tasked with distinct roles, while incorporating an obstacle avoidance function used to avoid both static and dynamic obstacles. Such results can be used in the applications of a high-capacity warehouse environments.

This study aimed to develop a real-time localization system for an AMR (autonomous mobile robot), which utilizes the Robot Operating System (ROS) Noetic version in the Ubuntu 20.04 operating system. RTAB-MAP (Real-Time Appearance-Based Mapping) is employed for localization, integrating with an RGB-D camera and a 2D LiDAR for real-time localization and mapping. The navigation was performed using the A* algorithm for global path planning, combined with the Dynamic Window Approach (DWA) for local path planning. It enables the AMR to receive velocity control commands and complete the navigation task. RTAB-MAP is a graph-based visual SLAM method that combines closed-loop detection and the graph optimization algorithm. The maps built using these three methods were evaluated with RTAB-MAP localization and AMCL (Adaptive Monte Carlo Localization) in a high-similarity long corridor environment. For RTAB-MAP and AMCL methods, three map optimization methods, i.e., TORO (Tree-based Network Optimizer), g2o (General Graph Optimization), and GTSAM (Georgia Tech Smoothing and Mapping), were used for the graph optimization of the RTAB-MAP and AMCL methods. Finally, the TORO, g2o, and GTSAM methods were compared to test the accuracy of localization for a long corridor according to the RGB-D camera and the 2D LiDAR.

Autonomous mobile robot (AMR) systems in indoor environments face challenges in jointly optimizing visual navigation and communication efficiency. Existing methods often decouple path planning from communication allocation or rely on simplified 2D models, limiting adaptability to 3D environments.. To address these limitations, we introduce the non-orthogonal multiple access technology to enhance spectral efficiency and ensure real-time control during AMR movement. We formulate a joint optimization problem to maximize the long-term average data rate via coordinated planning of AMR trajectories, transmit power, and precoding matrix. To handle wireless fading and localization inaccuracy, we propose a vision-enhanced federated deep reinforcement learning framework, where AMRs independently learn policies while engaging in privacy-preserving model aggregation. Leveraging RGB-D visual features, object detection, and egocentric pose estimation, our approach enables robust decision-making in 3D environments under imprecise localization. Extensive experiments on the RoboThor simulator demonstrate that our proposed framework accelerates training convergence, improves navigation and communication efficiency, and outperforms baselines in reward accumulation.

We propose a hierarchical dual-layer decision-making framework to address challenges associated with autonomous mobile robot (AMR) path planning in complex and dynamic campus environments. The upper-layer global planning is formulated as a multiobjective optimization model, where a multiobjective sheep flock migrate optimization algorithm (MOSFMO) is proposed to generate Pareto front solutions by optimizing path length and path safety jointly. In the lower layer, the autonomy of the AMR is enhanced through deep reinforcement learning (DRL) training with a composite reward scheme designed to enable resilient real-time decisions for avoiding unexpected pedestrians while achieving global objectives. Effective coordination is achieved through the availability of multiple candidate paths and a time-oriented deadlock detection mechanism, enabling uninterrupted task execution despite encountered blockage challenges. The proposed methods are validated through numerical simulations and real-world experiments, achieving on-time arrival rates of up to 99% in dynamic environments.

Current Automated Guided Vehicle (AGV) systems in warehouses often lack flexibility in navigating around obstacles due to limitation of line and marker-based navigation methods. This study presents an intelligent navigation system for an autonomous mobile robot (AMR) designed to improve efficiency of item retrieval in warehouse environment. The proposed system employs a Simultaneous Localization and Mapping (SLAM) algorithm to enable real-time mapping, obstacle avoidance, and efficient path generation. The global planner is a critical component in the navigation framework, as it determines the optimal and collision-free route through the warehouse, providing an efficient high-level path for the local planner to refine during real time navigation. Extensive evaluations were conducted to determine the optimal combination of global path planners between 1) RRT and Hybrid A* with local planners DWA and 2) TEB for path optimization, based on minimizing path length and travel time to the goal. Experimental results showed that the Hybrid A* with TEB planner achieved the fastest navigation time of 1.25 minutes with a 90% success rate, outperforming the RRT with DWA combination in both speed and reliability. This advancement offers significant potential to enhance warehouse operations by enabling fully autonomous mobile robots capable of performing item retrieval without human intervention.

Autonomous navigation of autonomous mobile robots (AMRs) in complex environments requires simultaneously addressing the validity of global paths and the optimization of local trajectories. This paper proposes a hierarchical AMR autonomous navigation framework that integrates smoothed rapidly-exploring random tree with soft actor-critic (SRRT-SAC), comprising an upper-level global motion planning layer and a lower-level local motion control layer. In the planning stage, we enhance the conventional rapid-exploring random tree (RRT) algorithm by applying B-spline interpolation to the globally planned path, thereby improving global path smoothness. In the control stage, the AMR navigation problem is formulated as a Markov decision process (MDP) and solved using deep reinforcement learning (DRL) with the Soft Actor-Critic (SAC) algorithm. We design a composite reward scheme that integrates heuristic functions to effectively mitigate the sparse-reward problem in DRL, and we apply dynamic normalization of the state space to accelerate and stabilize neural network training. Experimental results show that during training SRRT-SAC achieves an average reward convergence speed more than 36.8% faster than baseline methods such as SAC and TD3. In generalization tests, SRRT-SAC yields notable improvements over conventional DRL approaches across metrics including path length, task completion time, number of turns, and minimum obstacle clearance; notably, in a 20 m × 20 m environment it successfully completed navigation tasks that SAC, TD3, and A2C could not.

Aiming at the path optimization problem of multi-autonomous mobile robots (AMR/AGV) in cooperative sorting within a dynamic logistics environments, this paper proposes a path optimization framework based on multi-agent reinforcement learning (MARL). By constructing a state space that integrates local observation and the global environment, a cooperative strategy containing discrete and continuous actions is designed, and the real-time decision of AMR/AGV is optimized by multi-objective reward function, which provides theoretical support for intelligent sorting in large-scale logistics centers.

: This paper focuses on local dynamic path planning for autonomous vehicles, using an Adaptive Reinforcement Learning Twin Delayed Deep Deterministic Policy Gradient (ARL TD3) model. This model effectively navi-gates complex and unpredictable scenarios by adapting to changing environments. Testing, using simulations, shows improved path planning over static models, enhancing decision-making, trajectory optimization, and control. Challenges such as vehicle configuration, environmental factors, and top speed require further refine-ment. The model’s adaptability could be enhanced by integrating more data and exploring a fusion between supervised reinforcement learning and adaptive reinforcement learning techniques. This work advances autonomous vehicle path planning by introducing an ARL TD3 model for real-time decision-making in complex environments.

Path planning is a crucial component of autonomous mobile robot (AMR) systems. The slime mould algorithm (SMA), as one of the most popular path-planning approaches, shows excellent performance in the AMR field. Despite its advantages, there is still room for SMA to improve due to the lack of a mechanism for jumping out of local optimization. This means that there is still room for improvement in the path planning of ARM based on this method. To find shorter and more stable paths, an improved SMA, called the Lévy flight-rotation SMA (LRSMA), is proposed. LRSMA utilizes variable neighborhood Lévy flight and an individual rotation perturbation and variation mechanism to enhance the local optimization ability and prevent falling into local optimization. Experiments in varying environments demonstrate that the proposed algorithm can generate the ideal collision-free path with the shortest length, higher accuracy, and robust stability.

Abstract Planning a collision-free path in the least processing time and cost within constraints is a central issue in designing an autonomous mobile robot (AMR). Nature-inspired swarm intelligence (NISI) metaheuristic algorithms are gaining popularity in path planning and obstacle avoidance (PPOA) problem in AMRs. An efficient PPOA algorithm’s objective encompasses the ability to read a workspace map and consequently create the shortest collision-free path for the robot to maneuver from start to goal in the least processing time and effort. The authors have implemented a modified NISI metaheuristic approach known as a Slime Mould Optimization Algorithm (SMOA) in this research. SMOA takes inspiration from the oscillatory nature of slime mould when it encounters prey. Its mathematical model utilizes adaptive weights to simulate an optimal path for capturing prey or food. The slime moulds produce positive and negative feedback while propagating towards food with excellent exploratory competency and exploitation propensity. For this, simulation has been carried out on MATLAB 2020a. Additionally, the performance of SMOA has been compared with other NISI metaheuristic approaches such as PSO, FA, SFLA and ABC. The results demonstrate that modified SMOA takes less time and effort to generate an optimal collision-free path as compared to other mentioned approaches.

Planning a collision-free path while preserving processing time and minimizing cost function has been considered a significant challenge in developing an Autonomous Mobile Robot (AMR). Various optimization techniques for avoiding obstacles and path planning problems have been proposed recently. But, the computation time for executing these techniques is comparatively higher and has lesser accuracy. In this paper, the State Estimation Obstacle Avoidance (SEOA) algorithm has been proposed for estimating the position and velocity of both of the wheels of the AMR. Moreover, this algorithm has been also applied in path planning for reaching the destination point in minimum computational time. Five different positions of static obstacle are demonstrated in a real time static environment where the proposed SEOA algorithm has been compared with state-of-the-art path planning algorithms such as A* and VFH. The simulation results demonstrate that the proposed algorithm takes lesser computational time to generate the collision free path when compared to other mentioned algorithms.

With the development of factory automation and intelligent manufacturing system technology, an autonomous mobile robot (AMR) system has become an essential part of controlling the logistics management system within a facility. Relevant research about AMR path planning usually focuses on the fully autonomous environment that does not consider the uncertainty of human behavior. The behavior of human operators is unpredictable and therefore difficult to be integrated into the AMR system path planning analysis. In this paper, we propose an optimization algorithm for improving the pre-planned path considering the uncertainty of human behavior. Conditional value-at-risk constraints and chance constraints are considered in the optimization algorithm as the risk measurement to guarantee safety of operation. The performance of our approach is demonstrated through a 2-D AMR simulation, and the comparison of these two different risk measurements and their performance is also discussed.

This paper proposes a novel approach for finding an optimized solution for the online coverage path planning in unknown environments problem employing cooperative multi robotic agents. The suggested approach lessens the time required for solving the coverage problem by using cooperative multiple robotic agents working together to achieve enhanced complete coverage while sustaining minimum time, and the minimum number of repeated cells and respecting the service time. The distributed multi-agent planning with coordination process is optimized using dynamic programming with a rolling horizon limited look-ahead policy for online capability. In addition to expanding the usual single-agent action tree to hold a multi-agent scheme for possible collision-free sub-trajectories generation con-cerning quality, repetitions count, length per single agent to plan and execute cooperatively for covering a given unknown sur-roundings. Regarding coverage optimization, we utilize a multi-objective optimization genetic based algorithm with uniform cost search for the single-agent coverage sub-trajectory planning. In multi-agent-based systems, communication introduces overhead and have a direct influence on the system's robustness. Hence, we proposed a voting and coalition formation based on the K-means clustering algorithm for minimizing the communication henceforth improving the system's robustness. Simulation results highlight an enhancement regarding speeding up the complete coverage process compared to optimum single-agent approaches while advancing near-optimal utilization and efficiency minimizing repetitions in different unknown environments.

Traditional warehouse management systems face unprecedented challenges in the Industry 4.0 era, including escalating e-commerce demands, acute labor shortages, and critical requirements for real-time inventory visibility. Existing solutions fail to deliver the flexibility, scalability, and operational efficiency essential for contemporary supply chain operations. A novel integration framework combining Autonomous Mobile Robots (AMR) with Cyber-Physical Systems (CPS) is presented to enable intelligent, adaptive inventory management in smart warehouse environments. A multi-layered CPS architecture incorporating AMR fleet coordination, real-time data analytics, and digital twin synchronization is proposed. The framework employs distributed task allocation algorithms, dynamic path planning strategies, and predictive inventory optimization models. Implementation leverages edge computing for real-time decision-making and cloud infrastructure for comprehensive data analysis and storage. Experimental validation in industrial environments demonstrates significant performance improvements: 42% enhancement in order fulfillment speed, 35% reduction in inventory holding costs, and 89% accuracy in real-time stock tracking. The system maintained 99.2% uptime reliability while successfully managing 3× peak demand variations. The research advances smart logistics by establishing a scalable, generalizable CPS-AMR framework applicable across diverse warehouse environments. The findings provide actionable guidelines for Industry 4.0 transformation initiatives and establish theoretical foundations for next-generation autonomous warehouse systems.

The logistics industry is evolving rapidly, with an increasing demand for efficient and adaptive path planning strategies for Autonomous Mobile Robots (AMRs). This study evaluates three metaheuristic algorithms, Ant Colony Optimization (ACO), Genetic Algorithm (GA), and Sparrow Search Algorithm (SSA) for AMR path planning in a structured 20×20 grid-based environment. A comparative analysis is conducted based on execution time and path length. Results show that SSA achieved the shortest path length of 31.5 units, while GA had the fastest execution time of 1.74 seconds. ACO provided a balance between both metrics, achieving a path length of 32.8 units in 4.16 seconds. These findings contribute to optimizing AMR path planning strategies, improving smart logistics, and reducing operational costs.

This paper proposes an optimization framework that addresses both cycling degradation and calendar aging of batteries for autonomous mobile robot (AMR) to minimize battery degradation while ensuring task completion. A rectangle method of piecewise linear approximation is employed to linearize the bilinear optimization problem. We conduct a case study to validate the efficiency of the proposed framework in achieving an optimal path planning for AMRs while reducing battery aging.

Pure Pursuit (PP) is widely used in autonomous racing for real-time path tracking due to its efficiency and geometric clarity, yet performance is highly sensitive to how key parameters-lookahead distance and steering gain-are chosen. Standard velocity-based schedules adjust these only approximately and often fail to transfer across tracks and speed profiles. We propose a reinforcement-learning (RL) approach that jointly chooses the lookahead Ld and a steering gain g online using Proximal Policy Optimization (PPO). The policy observes compact state features (speed and curvature taps) and outputs (Ld, g) at each control step. Trained in F1TENTH Gym and deployed in a ROS 2 stack, the policy drives PP directly (with light smoothing) and requires no per-map retuning. Across simulation and real-car tests, the proposed RL-PP controller that jointly selects (Ld, g) consistently outperforms fixed-lookahead PP, velocity-scheduled adaptive PP, and an RL lookahead-only variant, and it also exceeds a kinematic MPC raceline tracker under our evaluated settings in lap time, path-tracking accuracy, and steering smoothness, demonstrating that policy-guided parameter tuning can reliably improve classical geometry-based control.

In recent years, wheeled mobile robotics (WMR) for small and large-scale Industry 4.0 applications are being implemented in warehouses, factories, and smart cities. Mobile robots must navigate constantly changing dynamic environments, which present significant challenges due to the difficulty of real-time mapping, collision avoidance, and path planning optimization. This research develops an autonomous mobile robot (AMR) system capable of navigating independently through unfamiliar and uncharted indoor environments. To achieve this, a sensor system tailored to the environment is used to perform specific tasks. The data collected by these sensors is processed by an enhanced SLAM (Simultaneous Localization and Mapping) algorithm, which extends SLAM's capabilities and generates pathways to unexplored regions. A simulation environment is created in Gazebo for mobile robot mapping, integrating lidar and odometry data throughout the process. The slip rate of the four-wheel robot's steering is measured in position, leading to improved chassis pose accuracy. Currently, ROS and STM32 communicate, with the ROS chassis node packaged to receive speed commands, provide feedback from odometer data, and process transformations. Based on experiments and simulations, the system accurately maps the environment and performs precise navigation tasks.

In recent years, land–air collaborative logistics delivery is a promising distribution method. It combines the flexibility of unmanned aerial vehicles (UAVs) with the high payload capacity of vehicles, expanding the service range of UAVs while reducing carbon emissions. However, most existing research has focused on path planning and resource allocation for either UAVs or vehicles alone. Therefore, to address the shortcomings of the current research, this paper proposes an intelligent land–air collaboration delivery algorithm for trajectory optimization and resource scheduling. Subsequently, this paper develops a land–air collaboration reliable delivery logistics distribution system, showcasing the driving routes of vehicles and UAVs. Meanwhile, the mode of UAV–vehicle collaboration not only saves operating costs compared to traditional logistics delivery but also achieves energy conservation and emission reduction goals. During the specific design and implementation process of this platform, blockchain technology is integrated into the logistics delivery service to ensure data security and prevent tampering, making the system more efficient and reliable. Finally, through testing and verification of the system’s functionalities, its completeness is demonstrated.

针对移动机器人完成特殊情况下的全覆盖路径规划(complete coverage path planning, CCPP)任务, 基于Lü系统, 提出一种构造混沌机器人的系统参数值综合选择策略, 以满足特殊任务下遍历轨迹高随机性和高覆盖率的需求。首先利用混沌系统必为耗散系统的特点, 大致确定Lü系统成为耗散系统的参数取值范围; 然后计算耗散系统下的李雅普诺夫指数谱, 缩小系统参数的取值范围; 其次画出这些参数下的相平面, 大致判断其轨迹的拓扑分布特性; 进一步在好的参数取值里, 计算每个参数下变量的皮尔逊相关系数, 判断每个变量的随机特性。最后, 在所确定参数值下, 利用其中的变量构造混沌机器人, 并仿真测试了覆盖率, 研究覆盖率和变量随机特性之间的关系。上述综合选择策略根据覆盖轨迹混沌性和随机性的要求, 逐渐缩小了系统参数的取值范围。与使用一组固定的经典参数值的Lü系统相比, 经过综合方法选择参数值的系统, 能挑选出李雅普诺夫指数大的变量来构造混沌机器人, 从而使覆盖轨迹的随机性能更高。另一个混沌Lorenz系统, 用来测试和验证所设计策略的可行性和有效性。此类研究能够提高机器人完成特殊情况下CCPP任务的效率。 We propose a novel parameter value selection strategy for the Lü system to construct a chaotic robot to accomplish the complete coverage path planning (CCPP) task. The algorithm can meet the requirements of high randomness and coverage rate to perform specific types of missions. First, we roughly determine the value range of the parameter of the Lü system to meet the requirement of being a dissipative system. Second, we calculate the Lyapunov exponents to narrow the value range further. Next, we draw the phase planes of the system to approximately judge the topological distribution characteristics of its trajectories. Furthermore, we calculate the Pearson correlation coefficient of the variable for those good ones to judge its random characteristics. Finally, we construct a chaotic robot using variables with the determined parameter values and simulate and test the coverage rate to study the relationship between the coverage rate and the random characteristics of the variables. The above selection strategy gradually narrows the value range of the system parameter according to the randomness requirement of the coverage trajectory. Using the proposed strategy, proper variables can be chosen with a larger Lyapunov exponent to construct a chaotic robot with a higher coverage rate. Another chaotic system, the Lorenz system, is used to verify the feasibility and effectiveness of the designed strategy. The proposed strategy for enhancing the coverage rate of the mobile robot can improve the efficiency of accomplishing CCPP tasks under specific types of missions.

No abstract available

No abstract available

No abstract available

This paper proposes a low-latency and scalable architecture for Edge-Enabled Digital Twin networked control systems (E-DTNCS) aimed at multi-robot collision avoidance and remote control in dynamic and latency-sensitive environments. Traditional approaches, which rely on centralized cloud processing or direct sensor-to-controller communication, are inherently limited by excessive network latency, bandwidth bottlenecks, and a lack of predictive decision-making, thus constraining their effectiveness in real-time multi-agent systems. To overcome these limitations, we propose a novel framework that seamlessly integrates edge computing with digital twin (DT) technology. By performing localized preprocessing at the edge, the system extracts semantically rich features from raw sensor data streams, reducing the transmission overhead of the original data. This shift from raw data to feature-based communication significantly alleviates network congestion and enhances system responsiveness. The DT layer leverages these extracted features to maintain high-fidelity synchronization with physical robots and to execute predictive models for proactive collision avoidance. To empirically validate the framework, a real-world testbed was developed, and extensive experiments were conducted with multiple mobile robots. The results revealed a substantial reduction in collision rates when DT was deployed, and further improvements were observed with E-DTNCS integration due to significantly reduced latency. These findings confirm the system’s enhanced responsiveness and its effectiveness in handling real-time control tasks. The proposed framework demonstrates the potential of combining edge intelligence with DT-driven control in advancing the reliability, scalability, and real-time performance of multi-robot systems for industrial automation and mission-critical cyber-physical applications.

Mobile robot teams often require decentralised autonomous navigation through narrow gaps in limited communication environments (e.g., underground search-and-rescue operations). Existing navigation approaches exhibit suboptimal performance for avoiding multi-robot collisions in such bottlenecks due to an inability to address the dynamic nature of the robots. Initial work utilising reinforcement learning has demonstrated success in navigating a single robot through narrow gaps. However, when training agents to produce give-way behaviour for navigating through constrained gaps, end-to-end reinforcement learning using simple rewards suffers from slow convergence due to the increased search space of viable policies. This paper introduces a novel curriculum reinforcement learning framework, incorporating a multi-robot bootstrap curriculum with preprogrammed behaviour to guide initial policy formation, subsequently refined by a gap curriculum that progressively reduces training complexity towards an optimal policy. This framework learns multi-robot interaction behaviours, which are impractical to program manually. Our model achieves a 99% success-rate in give-way behaviour generation without inter-agent communications in high-fidelity simulations. The success-rate reduced to 73% in simulations incorporating noisy sensors, and 60% in field-robot tests, substantiating our model's practical viability despite sensor noise and real-world uncertainties. The simple benchmark methods lack efficiency in basic interaction behaviours.

In this paper, we present a decentralized and communication-free collision avoidance approach for multi-robot systems that accounts for both robot localization and sensing uncertainties. The approach relies on the computation of an uncertainty-aware safe region for each robot to navigate among other robots and static obstacles in the environment, under the assumption of Gaussian-distributed uncertainty. In particular, at each time step, we construct a chance-constrained buffered uncertainty-aware Voronoi cell (B-UAVC) for each robot given a specified collision probability threshold. Probabilistic collision avoidance is achieved by constraining the motion of each robot to be within its corresponding B-UAVC, i.e. the collision probability between the robots and obstacles remains below the specified threshold. The proposed approach is decentralized, communication-free, scalable with the number of robots and robust to robots’ localization and sensing uncertainties. We applied the approach to single-integrator, double-integrator, differential-drive robots, and robots with general nonlinear dynamics. Extensive simulations and experiments with a team of ground vehicles, quadrotors, and heterogeneous robot teams are performed to analyze and validate the proposed approach.

Ensuring effective multi-robot collision avoidance in deep-sea environments presents unique challenges due to the absence of direct communication and the intermittency of positional updates. Existing methods struggle with outdated information, leading to suboptimal decision-making and increased risk of collisions. In this paper, we address these issues by introducing a multi-agent reinforcement learning (MARL)-based framework tailored for multi-robot navigation under intermittent observations. Our approach incorporates a recurrent neural network (RNN)-based module to take advantage of historical information, enabling agents to adapt to dynamic environments despite communication constraints. By leveraging MARL, our method effectively balances safety, efficiency, and path optimality, allowing robots to make informed decisions even with limited sensing updates. Simulations demonstrate that our approach significantly outperforms existing baselines, achieving higher success rates and improved adaptability in high-density, long-interval scenarios.

Multi-robot collision avoidance in decentralized manner is a challenging task which avoids collision with other robots and obstacles only based on its own observations. Recently, multi-agent reinforcement learning-based centralized training with decentralized execution has been widely used to train shared policy to reduce training cost. However, applying shared policies may limit the applicability in heterogeneous multi-robot systems. In this paper, we propose federated proximal policy algorithm (FedPPO) to share experience among robots with privacy preservation, which can be applied in heterogeneous robots. Each robot has its own policy and can upload the model parameter to server after several local training iterations. The server aggregates the model parameter to share experience. It is a capable solution addressing the low sampling efficiency of independent policy methods and the limited applicability of shared policy methods. The algorithms are trained on the irsim simulator. The experimental result shows that the proposed method has good sampling efficiency, and it can be used in heterogeneous multi-robot collision avoidance task.

Decentralized collision avoidance is a core challenge for scalable multi-robot systems. One of the promising approaches to tackle this problem is Model Predictive Path Integral (MPPI) -- a framework that naturally handles arbitrary motion models and provides strong theoretical guarantees. Still, in practice MPPI-based controller may provide suboptimal trajectories as its performance relies heavily on uninformed random sampling. In this work, we introduce CoRL-MPPI, a novel fusion of Cooperative Reinforcement Learning and MPPI to address this limitation. We train an action policy (approximated as deep neural network) in simulation that learns local cooperative collision avoidance behaviors. This learned policy is then embedded into the MPPI framework to guide its sampling distribution, biasing it towards more intelligent and cooperative actions. Notably, CoRL-MPPI preserves all the theoretical guarantees of regular MPPI. We evaluate our approach in dense, dynamic simulation environments against state-of-the-art baselines, such as ORCA, BVC, RL-RVO-NAV and classical MPPI. Our results demonstrate that CoRL-MPPI significantly improves navigation efficiency (measured by success rate and makespan) and safety, enabling agile and robust multi-robot navigation.

Multi-robot systems are increasingly being used for critical applications such as rescuing injured people, delivering food and medicines, and monitoring key areas. These applications usually involve navigating at high speeds through constrained spaces such as small gaps. Navigating such constrained spaces becomes particularly challenging when the space is crowded with multiple heterogeneous agents all of which have urgent priorities. What makes the problem even harder is that during an active response situation, roles and priorities can quickly change on a dime without informing the other agents. In order to complete missions in such environments, robots must not only be safe, but also agile, able to dodge and change course at a moment's notice. In this paper, we propose FACA, a fair and agile collision avoidance approach where robots coordinate their tasks by talking to each other via natural language (just as people do). In FACA, robots balance safety with agility via a novel artificial potential field algorithm that creates an automatic ``roundabout''effect whenever a conflict arises. Our experiments show that FACA achieves a improvement in efficiency, completing missions more than 3.5X faster than baselines with a time reduction of over 70% while maintaining robust safety margins.

Currently, precise spraying of sweet potatoes is mainly accomplished through semi-mechanized or single spraying robots, which results in low operating efficiency. Moreover, it is time-consuming and labor-intensive, and the pests and diseases cannot be eliminated in time. Based on multi robot navigation technology, multiple robots can work simultaneously, improving work efficiency. One of the main challenges faced by multi robot navigation technology is to develop a safe and robust collision avoidance strategy, so that each robot can safely and efficiently navigate from its starting position to the expected target. In this article, we propose a low-cost multi-robot collision avoidance method to solve the problem that multiple robots are prone to collision when working in field at the same time. This method has achieved good results in simulation. In particular, our collision avoidance method predicts the possibility of collision based on the robot’s position and environmental information, and changes the robot’s path in advance, instead of waiting for the robot to make a collision avoidance decision when it is closer. Finally, we demonstrate that a multi-robot collision avoidance approach provides an excellent solution for safe and effective autonomous navigation of a single robot working in complex sweet potato fields. Our collision avoidance method allows the robot to move forward effectively in the field without getting stuck. More importantly, this method does not require expensive hardware and computing power, nor does it require tedious parameter tuning.

Efficient coordination of multiple mobile robots is essential in automated systems, especially when robots must follow predefined paths while avoiding collisions. This paper proposes a centralized optimization framework using Genetic Algorithms to optimize the velocity profiles of a system of robots without altering their paths. The goal is to minimize task completion time and energy consumption while ensuring collision avoidance. Three Genetic Algorithm-based methods are introduced: Maximum Velocity Optimization, Slow-Down Segment Single-Objective Optimization and Slow-Down Segment Multi-Objective Optimization. The first method adjusts each robot’s maximum velocity along its entire path, whereas the second introduces a slow-down segment only at the start of its path. While these two approaches only optimize task completion time, the third method contains a multi-objective formulation, producing solutions that balance time and energy. Methods such as Brute-Force and Prioritized Planning were used as baseline methods for comparison. Simulation results indicate that the proposed strategies significantly outperform the baseline methods. Furthermore, the second method achieves better results than the first by introducing more targeted velocity adjustments, while the third further enhances flexibility by offering a range of trade-offs between task completion time and energy consumption. Scalability and computational cost remain critical challenges, especially as the number of robots increases.

The Iterative Forecast Planner (IFP) is a geometric planning approach that offers lightweight computations, scalable, and reactive solutions for multi-robot path planning in decentralized, communication-free settings. However, it struggles in symmetric configurations, where mirrored interactions often lead to collisions and deadlocks. We introduce eIFP-MPC, an optimized and extended version of IFP that improves robustness and path consistency in dense, dynamic environments. The method refines threat prioritization using a time-to-collision heuristic, stabilizes path generation through cost-based viapoint selection, and ensures dynamic feasibility by incorporating model predictive control (MPC) into the planning process. These enhancements are tightly integrated into the IFP to preserve its efficiency while improving its adaptability and stability. Extensive simulations across symmetric and high-density scenarios show that eIFP-MPC significantly reduces oscillations, ensures collision-free motion, and improves trajectory efficiency. The results demonstrate that geometric planners can be strengthened through optimization, enabling robust performance at scale in complex multi-agent environments.

Deep reinforcement learning has been demonstrated to be an effective solution to the multi-robot collision avoidance problem. However, with existing methods, robots typically generate actions only based on local observations, sometimes augmented with global communication. Their performance deteriorates in limited bandwidth environments and complex scenarios with various obstacles and high robot density. We propose SelComm, a selective communication framework to generate cooperative and collision-free actions for robots in multi-robot navigation tasks. Specifically, we develop a decentralized message selector, enabling each robot to calculate relations with other robots using both agent-level information and sensor-level information, and select the most valuable messages to meet the bandwidth limitation. Then we introduce the attentional communication channel for efficient communication. Our experimental evaluations based on various scenarios demonstrate that SelComm learns more cooperative behaviors and outperforms state-of-the-art methods in limited bandwidth environments and complex scenarios.

This paper introduces Chance-Constrained Safety Barrier Certificates (CC-SBC) for decentralized collision avoidance of multi-robot systems that account for uncertainty in robot state estimation. The work builds upon the previously proposed Safety Barrier Certificates (SBC) approach. We present a probabilistic formulation by introducing CC-SBC, a chance-constrained set that defines a probabilistic safe control space for each robot in the system. The CC-SBC chance constraints are then reformulated into a set of deterministic quadratic constraints, which allows the robot to find a safe controller by solving a quadratically constrained quadratic program (QCQP). We validate through simulations that the proposed approach can achieve safe navigation of a team of robots taking into account the robots' state estimation uncertainties.

This paper presents a novel decentralized multi-robot collision avoidance method with deep reinforcement learning, which is not only suitable for the large-scale grid map workspace multi-robot system, but also directly processes Lidar signals instead of communicating between the robots. According to the particularity of the workspace, we handcrafted a reward function, which considers both the collision avoidance among the robots and as little as possible change of direction of the robots during driving. Using Double Deep Q-Network (DDQN), the policy was trained in the simulation grid map workspace. By designing experiments, we demonstrated that the learned policy can guide the robot well to effectively travel from the initial position to the goal position in the grid map workspace and to avoid collisions with others while driving.

This paper proposes an online path planning method for multi-robot collision avoidance that accounts for uncertainties in robot positioning. Firstly, the online path plan-ning problem for multiple robots is transformed into a multi-objective optimization problem by considering some necessary factors that affect task efficiency and safety. Notably, the paper considers the uncertainty in robot positions, in which the position follows a Gaussian distribution. Then, by converting probabilistic collision conditions into deterministic constraints, the fitness function ensuring probabilistic collision avoidance is constructed. Next, an improved sine-cosine algorithm is proposed to solve the ideal position. Finally, simulation experiments demonstrate that the proposed method effectively resolves collision issues under probabilistic uncertainty conditions.

No abstract available

Optimal path planning has always been a key issue in multi-robot research. One of the most important issues is how to use appropriate methods to resolve path collision between multiple robots, which is usually converted into an optimization problem of time and space. This paper solves the problem of path collision at intersections between robots based on the change of RSSI value. First, a fixed priority collision avoidance algorithm based on the robot number is proposed. In order to improve this algorithm, a binary sequence priority collision avoidance algorithm is proposed. In Webots Robot Simulator, self-build robots are used to implement the proposed algorithm.

This paper designs a motion rule suitable for grid environment and proposes a multi-robot autonomous obstacle avoidance method based on deep reinforcement learning. The training and validation of the method are done under the Stage simulation platform based on ROS operating system. During the training process, the robot uses Lidar to obtain the surrounding state information and generates actions based on the state information to obtain rewards, and the robot is guided by the rewards to optimize the strategy. Based on the D3QN algorithm, a new reward function is designed, a proximity penalty is introduced to reduce the collision between robots, a distance reward is added to guide the robot to complete the task, a step reward is added to improve the efficiency of the robot to complete the task, and an illegal action penalty is added to avoid the robot to choose an illegal action; the input is 5 frames of Lidar data, and in the network structure, the agent can better learn the correlation between the data by introducing Long Short Term Memory(LSTM) layer, and introducing Convolutional Block Attention Module(CBAM), a hybrid attention mechanism to allow the robot to pay more attention to the information of the surrounding robots. By designing experiments, we demonstrate that the learned strategy can effectively guide the robot through obstacle avoidance and complete the task.

Collision avoidance is a fundamental problem in multi-robot motion planning, enabling robots to coordinate in a safe and efficient manner. In this paper, we introduce several best-known distributed collision avoidance algorithms and present the Voronoi-based method in detail, which assigns optimal dominance regions for robots using the Voronoi diagram, constrains robot motions within the buffered Voronoi diagram, and follows right-hand rule when deadlock happens during navigating to the goals. A series of simulations are conducted to demonstrate the efficacy of collision avoidance guarantee under robot localization uncertainty while avoiding deadlock using our presented algorithm. We claim that such a method is superior to most of the existing distribution methods in a known environment in efficiency of implementation.

Multi-robot path planning is an essential research topic for coordinated robots. In this paper, we aim to enable teams of robots to navigate in a common environment samultanuously without collision. Our method is based on Voronoi diagram, an optimal spatial area assignment strategy. We take the localization inaccuracy in to consideration, which always exist in real-world senarioss and set a buffer to the ordinary Voronoi diagram for collision-free navigation amoung the robot team. We also combine the left-handed principle to handle the situation of deadlock, which may happen when robots reach the connors or edges of their Voronoi cells. We use MATLAB simulation to test the efficacy of our proposed algorithms in an open environment with twelve mobile robots. The results show that all robots reach the target efficiently without collision. We also compare our method with several related multi-robot path planning algorithms to demonstrate the efficiency of our proposed method.

In this paper, a novel framework is presented that achieves a combined solution based on Multi-Robot Task Allocation (MRTA) and collision avoidance with respect to homogeneous measurement tasks taking place in industrial environments. The spatial clustering we propose offers to simultaneously solve the task allocation problem and deal with collision risks by cutting the workspace into distinguishable operational zones for each robot. To divide task sites and to schedule robot routes within corresponding clusters, we use K-means clustering and the 2-Opt algorithm. The presented framework shows satisfactory performance, where up to 93\% time reduction (1.24s against 17.62s) with a solution quality improvement of up to 7\% compared to the best performing method is demonstrated. Our method also completely eliminates collision points that persist in comparative methods in a most significant sense. Theoretical analysis agrees with the claim that spatial partitioning unifies the apparently disjoint tasks allocation and collision avoidance problems under conditions of many identical tasks to be distributed over sparse geographical areas. Ultimately, the findings in this work are of substantial importance for real world applications where both computational efficiency and operation free from collisions is of paramount importance.

Recent research topics have been placed on reinforcing security and safety measures within high-risk industries to protect both equipment and the environment. Numerous industries carry substantial implications for their surroundings. During significant incidents like chemical spills, or nuclear accidents, swiftly gathering precise and dynamic data poses a considerable challenge. Subsequently, this paper focuses on optimizing a mission involving multiple mobile robots charged with inspecting an industrial zone. The aim of this research is to efficiently collect measurements from diverse positions using a fleet of sensing robots operating from a central depot, that is, developing an algorithm for robot decision-making that optimizes mission planning by minimizing an objective function. Initially, we explore our previous proposed solutions and we improve the system by integrating a navigational layer to manage collision avoidance between robots. Then, we delve into scenarios involving multiple homogeneous tasks distributed in a limited geographical environment. To demonstrate feasibility, extensive simulations, numerical experiments, and comparative analysis are conducted, showing the efficiency of the proposed approaches in terms of solution quality and computational complexity.

: This study proposes integrating Reciprocal Velocity Obstacles (RVO) with Bare Bones Particle Swarm Optimization (BB-PSO) for prioritized motion planning in multi-robot systems. BB-PSO was chosen because it has fewer parameters to tune, reduced computational complexity, and provides potentially faster convergence compared to standard PSO. The methodology enables collision avoidance and path planning while allowing differentiated robot behaviors based on priority levels. Simulations used a two-phase experimental strategy: first, tuning cost function parameters through grid search, and second, evaluating various priority configurations and random scenarios. Results show that the selected weight configuration ( α = 4 , β = 2) balances goal-seeking and obstacle avoidance, enabling high-priority agents to move directly while ensuring overall group safety. Scenarios with higher average priorities exhibited shorter travel distances and faster completion times, whereas those with lower or imbalanced priorities led to more conservative behavior and delays. Compared to a greedy baseline, the proposed method significantly reduced collisions, achieving an average of 1.0 collision per scenario versus 6.6 with the greedy approach. Some priority configurations achieved complete task fulfillment without any collisions, highlighting the potential for optimized multi-robot coordination. The proposed method offers a promising strategy for prioritized motion planning, balancing efficiency and safety based on task importance. Future research includes comparing BB-PSO with other optimization methods, reducing sample requirements, dynamically adjusting priorities, and extending the model to incorporate task parameterizations and autonomous priority adaptation.

This article is dedicated to solving the obstacle avoidance problem of multi-robots for two-wheel differential robots. The Kinematics constraints of the two-wheeled robot are introduced based on the uncertainty models for TTC-Based Collision Avoidance (UTTC), and the restricted UTTC (RUTTC) model is obtained. In the RUTTC, robots have smoother changes in orientation angle, less crowding, faster speed, and shorter running time. Further, the greedy method will be utilized to improve the path planned by PRM, guiding RUTTC away from local minima. Ultimately, multiple two-wheeled differential robots can avoid each other and reach the target point without colliding with obstacles in more complex environments.

A multi-robot navigation and obstacle avoidance algorithm is proposed based on multi-agent deep deterministic policy gradient (MADDPG) framework. Compared to single-agent reinforcement learning algorithms, shared information among multiple agents is leveraged in the proposed approach, significantly enhancing the model's convergence speed and stability. To address the inefficiencies caused by sparse rewards, an artificial potential field (APF) algorithm is integrated to guide agents in navigating obstacles and efficiently reaching designated targets. An adaptive APF algorithm is designed in this study, through which agents' motion performance is improved. Besides, a novel prioritized experience replay algorithm is proposed, which addresses the shortcoming of the traditional method, where the sampling probability depends solely on the TD-error. From simulation results, the proposed methods are demonstrated to allow agents to avoid obstacles smoothly and reach their targets, achieving a higher success rate in obstacle avoidance and faster navigation than traditional algorithms.

Control barrier function (CBF)-based methods provide the minimum modification necessary to formally guarantee safety in the context of quadratic programming, and strict safety guarantee for safety critical systems. However, most CBF-related derivatives myopically focus on present safety at each time step, a reasoning over a look-ahead horizon is exactly missing. In this paper, a predictive safety matrix is constructed. We then consolidate the safety condition based on the smallest eigenvalue of the proposed safety matrix. A predefined deconfliction strategy of motion paths is embedded into the trajectory tracking module to manage deadlock conflicts, which computes the deadlock escape velocity with the minimum attitude angle. Comparison results show that the introduction of the predictive term is robust for measurement uncertainty and is immune to oscillations. The proposed deadlock avoidance method avoids a large detour, without obvious stagnation.

Collision avoidance with moving obstacles is a fundamental problem for multi-robot systems. This paper presents a new collision avoidance control method involving moving obstacles without solving a numerical optimization problem. When the current relative velocity between a robot and a moving obstacle is close to or within the collision cone, the heading angle of a robot is modified such that the direction of the relative velocity with a moving obstacle moves away from the collision cone. Constraints on the heading angular velocity of robots based on a control barrier function are utilized to prevent the relative velocity with obstacles from entering the collision cone. This is the main distinction from previous methods based on collision cones. Furthermore, collision avoidance is verified by validating position-based constraints on the computed control input. The effectiveness of the proposed method is demonstrated through simulations and experiments.

No abstract available

No abstract available

This work focuses on the solution for multi-agent, collision-free, time-varying path-following problems in 3D spaces. The proposed solution is based on using time-varying artificial vector fields to generate velocity references for single agents. A distributed Model Predictive Control (MPC) scheme is employed, taking into account the system’s dynamics and collision avoidance features, enabling the multi-robot system to follow the dynamic field without collisions and providing great computational efficiency. More specifically, Optimal Reciprocal Collision Avoidance (ORCA) algorithm is used to set constraints, yielding a novel MPC-ORCA approach, for efficient collision avoidance in the multi-agent scenario. Simulation scenarios are used to validate and compare the proposed approach with traditional methods, highlighting its improvements.

Multi-agent time-varying path following problems still offer a wide variety of open challenges, in which efficient collision avoidance is of great importance in this context. This work proposes a solution based on artificial vector fields that generate velocity references for single agents in path-following tasks. A distributed Model Predictive Control (MPC) scheme accountable for double integrator dynamic models and collision avoidance features enables the group of robots to follow the dynamic field in a safe manner. Control Barrier Functions (CBF) are utilized to include collision avoidance in the MPC problem. Simulation scenarios corroborate the method’s efficiency and highlight the improvements in contrast with previous works.

Obstacle avoidance for multi-robot navigation with polytopic shapes is challenging. Existing works simplify the system dynamics or consider it as a convex or non-convex optimization problem with positive distance constraints between robots, which limits real-time performance and scalability. Additionally, generating collision-free behavior for polytopic-shaped robots is harder due to implicit and non-differentiable distance functions between polytopes. In this letter, we extend the concept of velocity obstacle (VO) principle for polytopic-shaped robots and propose a novel approach to construct the VO in the function of vertex coordinates and other robot's states. Compared with existing work about obstacle avoidance between polytopic-shaped robots, our approach is much more computationally efficient as the proposed approach for construction of VO between polytopes is optimization-free. Based on VO representation for polytopic shapes, we later propose a navigation approach for distributed multi-robot systems. We validate our proposed VO representation and navigation approach in multiple challenging scenarios including large-scale randomized tests, and our approach outperforms the state of art in many evaluation metrics, including completion rate, deadlock rate, and the average travel distance.

Multi-robot systems are experiencing increasing popularity in joint rescue, intelligent transportation, and other fields. However, path planning and navigation obstacle avoidance among multiple robots, as well as dynamic environments, raise significant challenges. We propose a distributed multi-mobile robot navigation and obstacle avoidance method in unknown environments. First, we propose a bidirectional alternating jump point search A* algorithm (BAJPSA*) to obtain the robot's global path in the prior environment and further improve the heuristic function to enhance efficiency. We construct a robot kinematic model based on the dynamic window approach (DWA), present an adaptive navigation strategy, and introduce a new path tracking evaluation function that improves path tracking accuracy and optimality. To strengthen the security of obstacle avoidance, we modify the decision rules and obstacle avoidance rules of the single robot and further improve the decision avoidance capability of multi-robot systems. Moreover, the mainstream prioritization method is used to coordinate the local dynamic path planning of our multi-robot systems to resolve collision conflicts, reducing the difficulty of obstacle avoidance and simplifying the algorithm. Experimental results show that this distributed multi-mobile robot motion planning method can provide better navigation and obstacle avoidance strategies in complex dynamic environments, which provides a technical reference in practical situations.

Real-time dynamic path planning in complex traffic environments presents challenges, such as varying traffic volumes and signal wait times. Traditional static routing algorithms like Dijkstra and A* compute shortest paths but often fail under dynamic conditions. Recent Reinforcement Learning (RL) approaches offer improvements but tend to focus on local optima, risking dead-ends or boundary issues. This paper proposes a novel approach based on causal inference for real-time dynamic path planning, balancing global and local optimality. We first use the static Dijkstra algorithm to compute a globally optimal baseline path. A distributed control strategy then guides vehicles along this path. At intersections, DynamicRouteGPT performs real-time decision-making for local path selection, considering real-time traffic, driving preferences, and unexpected events. DynamicRouteGPT integrates Markov chains, Bayesian inference, and large-scale pretrained language models like Llama3 8B to provide an efficient path planning solution. It dynamically adjusts to traffic scenarios and driver preferences and requires no pre-training, offering broad applicability across road networks. A key innovation is the construction of causal graphs for counterfactual reasoning, optimizing path decisions. Experimental results show that our method achieves state-of-the-art performance in real-time dynamic path planning for multiple vehicles while providing explainable path selections, offering a novel and efficient solution for complex traffic environments.

With the rapid growth of digital cities, massive sensing data distributed along road segments and roadside infrastructure need to be acquired and processed in a timely manner to support intelligent operations and infrastructure management. Owing to recent developments in unmanned ground vehicles (UGVs) technology and edge computing, UGVs can offload tasks to roadside units (RSUs), thereby greatly enhancing their data acquisition and processing capabilities. In this work, we consider both green-energy and grid-energy supplied RSUs. We deploy the UGV to traverse all target road segments in the considered road network, ensuring comprehensive acquisition of all data distributed along the roads. Furthermore, we introduce a composite reward to capture both the sustainability and time-sensitivity during data acquisition. Then, we establish a UGV reward collection problem that jointly optimizes UGV routing, speed control, task offloading ratios, and energy management. Considering the complexity of the problem, we propose an efficient iterative algorithm to solve it, in which an improved Branch-and-Bound algorithm is derived. Numerical experiments demonstrate that our approach outperforms baseline methods, while also effectively balancing timeliness and green-energy utilization.

Unmanned ground vehicles (UGVs) require effective perception and analysis of their surrounding terrain for safe operation. This paper presents a novel approach to their local elevation mapping and traversability analysis using sparse data from a single LiDAR sensor, which can generate a dense local traversability map in real-time. By preserving ground height information, our method can differentiate between vertical obstacles, suspended objects and other terrains in the elevation map. The modified Bayesian generalized kernel elevation inference is utilized to predict and fill in sparse elevation maps to achieve local dense terrain traversability mapping. The system uses GPU parallel processing to accelerate calculations, ensuring real-time perception and dynamic processing. The proposed system was tested in both structured and unstructured environments, and achieved better performances in map filling and handling of suspended and vertical objects compared to other existing approaches.

The advent of autonomous vehicles (AVs) marks a significant milestone in urban transportation, promising to enhance safety, reduce congestion, and improve environmental sustainability. However, deploying AVs on a mass scale comes with critical challenges related to secure and efficient vehicular communication. This research work proposes a novel framework that combines the security features of blockchain technology with the adaptive capabilities of machine learning (ML) to address these major challenges. Integrating a blockchain-based protocol ensures tamper-proof and transparent communication within AV networks, protecting against a wide array of cyber threats. Concurrently, ML algorithms are employed to optimize real-time routing decisions based on comprehensive traffic data and environmental conditions. Through simulation in realistic urban scenarios, our framework demonstrates a significant improvement in communication security and routing efficiency, indicating a promising avenue for achieving scalable and reliable AV networks. Operational cost assessments further reveal the economic viability of the proposed model, underscoring its potential to deliver long-term savings through enhanced efficiency and reduced human intervention. Thus an efficient solution in terms of security, dynamic routing, and scalability with respect to traditional models.

In this paper, a new Messenger Routing Problem (MRP) is studied, which is motivated by Uncrewed Aerial Vehicles (UAVs) accessing Uncrewed Ground Vehicles (UGVs) to deliver information in UAV-UGV coordination systems. The objective is to minimize the longest path among multiple messengers, ensuring fast and reliable information transmission. Two key challenges arise in tackling this problem. First, the targets are moving, incurring the travel cost between targets to vary with the travel process. Second, the messengers accessing the neighborhood of targets needs to satisfy the communication time constraint. Based on the idea of decoupling, a two-phase planner is proposed to sequentially determine global access sequence and optimize local access locations. In the first phase, a motion prediction module is introduced in the Adaptive Large Neighborhood Search (ALNS) framework to deal with the dynamic characteristics of MRP. In the second phase, an efficient bisection sampling method based on the prediction points is proposed to obtain a shorter access path while satisfying the communication time constraint. Finally, the effectiveness and efficiency of the proposed method are demonstrated by performance evaluation and comparison with the state-of-the-art algorithms. Note to Practitioners—This paper studies an emerging UAV routing problem, the Messenger Routing Problem, which is aimed at providing communication services for multiple mobile UGVs. These UGVs are tasked with large-scale operations such as firefighting, environmental monitoring, and disaster relief, where ground communication is often limited or unreliable. Due to their constrained communication capabilities, UGVs may struggle to obtain timely and essential environmental information. By employing UAVs as messengers to detect the scene and fly over the effective communication range of all mobile UGVs to deliver information, a cooperative air-ground network can be established, which allows for successful operation in environments with unavailable communication infrastructure and poor network connectivity, enabling large-scale collaboration among multiple UGVs. This paper proposes a two-phase planner to effectively search for UAV routes to deliver information in the shortest time. The planner first determines the global access sequence and then optimizes the local access locations. Simulation results demonstrate that the proposed method can generate high-quality tours for messenger UAVs effectively and efficiently. We also offer operational insights for real-world deployments and outline strategies for optimizing the system under hardware constraints. In future work, we aim to focus more on real-world scenarios involving uncertain UGV motion patterns.

The cooperative control of the unmanned ground vehicle (UGV) formation is crucial for improving the efficiency and safety of intelligent transportation systems. Considering their complex and dynamic operating environments, the scene-adaptive switching control algorithm is proposed to address this challenge. Firstly, for the obstacle avoidance of a single UGV within a formation, a dynamic window approach that can adjust the weight of the formation function in real time is presented. By minimizing the deviation between the desired formation position and the actual position, a single UGV can quickly recover the original formation after completing local obstacle avoidance. For the cooperative obstacle avoidance of multiple UGVs, a multi-layer hierarchical obstacle avoidance strategy is designed, which is based on the constraints such as obstacle threat level and density. Furthermore, an improved consensus algorithm is used to achieve rapid reconfiguration of the formation, and a target factor is introduced into the leader-follower algorithm to ensure that the UGVs reach the target position smoothly. Finally, the simulation and experimental results show that the multi-vehicle system can dynamically avoid obstacles, quickly maintain formation, and reconfigure formation.

Approximate computing is a computation domain which can be used to trade time and energy with quality and therefore is useful in embedded systems. Energy is the prime resource in battery-driven embedded systems, like robots. Approximate computing can be used as a technique to generate approximate version of the control functionalities of a robot, enabling it to ration energy for computation at the cost of degraded quality. Usually, the programmer of the function specifies the extent of degradation that is safe for the overall safety of the system. However, in a collaborative environment, where several sub-systems co-exist and some of the functionality of each of them have been approximated, the safety of the overall system may be compromised. In this paper, we consider multiple identical robots operate in a warehouse, and the path planning function of the robot is approximated. Although the planned paths are safe for individual robots (i.e. they do not collide with the racks), we show that this leads to a collision among the robots. So, a controlled approximation needs to be carried out in such situations to harness the full power of this new paradigm if it needs to be a mainstream paradigm in future.

Considering an environment containing polygonal obstacles, we address the problem of planning motions for a pair of planar robots connected to one another via a cable of limited length. Much like prior problems with a single robot connected via a cable to a fixed base, straight line-of-sight visibility plays an important role. The present paper shows how the reduced visibility graph provides a natural discretization and captures the essential topological considerations very effectively for the two robot case as well. Unlike the single robot case, however, the bounded cable length introduces considerations around coordination (or equivalently, when viewed from the point of view of a centralized planner, relative timing) that complicates the matter. Indeed, the paper has to introduce a rather more involved formalization than prior single-robot work in order to establish the core theoretical result -- a theorem permitting the problem to be cast as one of finding paths rather than trajectories. Once affirmed, the planning problem reduces to a straightforward graph search with an elegant representation of the connecting cable, demanding only a few extra ancillary checks that ensure sufficiency of cable to guarantee feasibility of the solution. We describe our implementation of A${}^\star$ search, and report experimental results. Lastly, we prescribe an optimal execution for the solutions provided by the algorithm.

Trajectory planning for multiple robots in shared environments is a challenging problem especially when there is limited communication available or no central entity. In this article, we present Real-time planning using Linear Spatial Separations, or RLSS: a real-time decentralized trajectory planning algorithm for cooperative multi-robot teams in static environments. The algorithm requires relatively few robot capabilities, namely sensing the positions of robots and obstacles without higher-order derivatives and the ability of distinguishing robots from obstacles. There is no communication requirement and the robots' dynamic limits are taken into account. RLSS generates and solves convex quadratic optimization problems that are kinematically feasible and guarantees collision avoidance if the resulting problems are feasible. We demonstrate the algorithm's performance in real-time in simulations and on physical robots. We compare RLSS to two state-of-the-art planners and show empirically that RLSS does avoid deadlocks and collisions in forest-like and maze-like environments, significantly improving prior work, which result in collisions and deadlocks in such environments.

Mobile manipulators have been employed in many applications that are traditionally performed by either multiple fixed-base robots or a large robotic system. This capability is enabled by the mobility of the mobile base. However, the mobile base also brings redundancy to the system, which makes mobile manipulator motion planning more challenging. In this paper, we tackle the mobile manipulator motion planning problem under the end-effector trajectory continuity constraint in which the end-effector is required to traverse a continuous task-space trajectory (time-parametrized path), such as in mobile printing or spraying applications. Our method decouples the problem into: (1) planning an optimal base trajectory subject to geometric task constraints, end-effector trajectory continuity constraint, collision avoidance, and base velocity constraint; which ensures that (2) a manipulator trajectory is computed subsequently based on the obtained base trajectory. To validate our method, we propose a discrete optimal base trajectory planning algorithm to solve several mobile printing tasks in hardware experiment and simulations.

Modern robotic manufacturing requires collision-free coordination of multiple robots to complete numerous tasks in shared, obstacle-rich workspaces. Although individual tasks may be simple in isolation, automated joint task allocation, scheduling, and motion planning under spatio-temporal constraints remain computationally intractable for classical methods at real-world scales. Existing multi-arm systems deployed in the industry rely on human intuition and experience to design feasible trajectories manually in a labor-intensive process. To address this challenge, we propose a reinforcement learning (RL) framework to achieve automated task and motion planning, tested in an obstacle-rich environment with eight robots performing 40 reaching tasks in a shared workspace, where any robot can perform any task in any order. Our approach builds on a graph neural network (GNN) policy trained via RL on procedurally-generated environments with diverse obstacle layouts, robot configurations, and task distributions. It employs a graph representation of scenes and a graph policy neural network trained through reinforcement learning to generate trajectories of multiple robots, jointly solving the sub-problems of task allocation, scheduling, and motion planning. Trained on large randomly generated task sets in simulation, our policy generalizes zero-shot to unseen settings with varying robot placements, obstacle geometries, and task poses. We further demonstrate that the high-speed capability of our solution enables its use in workcell layout optimization, improving solution times. The speed and scalability of our planner also open the door to new capabilities such as fault-tolerant planning and online perception-based re-planning, where rapid adaptation to dynamic task sets is required.

With the recent influx in demand for multi-robot systems throughout industry and academia, there is an increasing need for faster, robust, and generalizable path planning algorithms. Similarly, given the inherent connection between control algorithms and multi-robot path planners, there is in turn an increased demand for fast, efficient, and robust controllers. We propose a scalable joint path planning and control algorithm for multi-robot systems with constrained behaviours based on factor graph optimization. We demonstrate our algorithm on a series of hardware and simulated experiments. Our algorithm is consistently able to recover from disturbances and avoid obstacles while outperforming state-of-the-art methods in optimization time, path deviation, and inter-robot errors. See the code and supplementary video for experiments.

Trajectory Planning is a crucial word in Modern & Advanced Robotics. It's a way of generating a smooth and feasible path for the robot to follow over time. The process primarily takes several factors to generate the path, such as velocity, acceleration and jerk. The process deals with how the robot can follow a desired motion path in a suitable environment. This trajectory planning is extensively used in Automobile Industrial Robot, Manipulators, and Mobile Robots. Trajectory planning is a fundamental component of motion control systems. To perform tasks like pick and place operations, assembly, welding, painting, path following, and obstacle avoidance. This paper introduces a comparative analysis of trajectory planning algorithms and their key software elements working strategy in complex and dynamic environments. Adaptability and real-time analysis are the most common problems in trajectory planning. The paper primarily focuses on getting a better understanding of these unpredictable environments.

Path planning is an important problem with the the applications in many aspects, such as video games, robotics etc. This paper proposes a novel method to address the problem of Deep Reinforcement Learning (DRL) based path planning for a mobile robot. We design DRL-based algorithms, including reward functions, and parameter optimization, to avoid time-consuming work in a 2D environment. We also designed an Two-way search hybrid A* algorithm to improve the quality of local path planning. We transferred the designed algorithm to a simple embedded environment to test the computational load of the algorithm when running on a mobile robot. Experiments show that when deployed on a robot platform, the DRL-based algorithm in this article can achieve better planning results and consume less computing resources.

Mobility on asteroids by multi-limbed climbing robots is expected to achieve our exploration goals in such challenging environments. We propose a mobility strategy to improve the locomotion safety of climbing robots in such harsh environments that picture extremely low gravity and highly uneven terrain. Our method plans the gait by decoupling the base and limbs' movements and adjusting the main body pose to avoid ground collisions. The proposed approach includes a motion planning that reduces the reactions generated by the robot's movement by optimizing the swinging trajectory and distributing the momentum. Lower motion reactions decrease the pulling forces on the grippers, avoiding the slippage and flotation of the robot. Dynamic simulations and experiments demonstrate that the proposed method could improve the robot's mobility on the surface of asteroids.

This paper aims to improve the path quality and computational efficiency of kinodynamic planners used for vehicular systems. It proposes a learning framework for identifying promising controls during the expansion process of sampling-based motion planners for systems with dynamics. Offline, the learning process is trained to return the highest-quality control that reaches a local goal state (i.e., a waypoint) in the absence of obstacles from an input difference vector between its current state and a local goal state. The data generation scheme provides bounds on the target dispersion and uses state space pruning to ensure high-quality controls. By focusing on the system's dynamics, this process is data efficient and takes place once for a dynamical system, so that it can be used for different environments with modular expansion functions. This work integrates the proposed learning process with a) an exploratory expansion function that generates waypoints with biased coverage over the reachable space, and b) proposes an exploitative expansion function for mobile robots, which generates waypoints using medial axis information. This paper evaluates the learning process and the corresponding planners for a first and second-order differential drive systems. The results show that the proposed integration of learning and planning can produce better quality paths than kinodynamic planning with random controls in fewer iterations and computation time.

Autonomous mobile robots are usually faced with challenging situations when driving in complex environments. Namely, they have to recognize the static and dynamic obstacles, plan the driving path and execute their motion. For addressing the issue of perception and path planning, in this paper, we introduce OctoPath , which is an encoder-decoder deep neural network, trained in a self-supervised manner to predict the local optimal trajectory for the ego-vehicle. Using the discretization provided by a 3D octree environment model, our approach reformulates trajectory prediction as a classification problem with a configurable resolution. During training, OctoPath minimizes the error between the predicted and the manually driven trajectories in a given training dataset. This allows us to avoid the pitfall of regression-based trajectory estimation, in which there is an infinite state space for the output trajectory points. Environment sensing is performed using a 40-channel mechanical LiDAR sensor, fused with an inertial measurement unit and wheels odometry for state estimation. The experiments are performed both in simulation and real-life, using our own developed GridSim simulator and RovisLab's Autonomous Mobile Test Unit platform. We evaluate the predictions of OctoPath in different driving scenarios, both indoor and outdoor, while benchmarking our system against a baseline hybrid A-Star algorithm and a regression-based supervised learning method, as well as against a CNN learning-based optimal path planning method.

We present CLIPSwarm, an algorithm to generate robot swarm formations from natural language descriptions. CLIPSwarm receives an input text and finds the position of the robots to form a shape that corresponds to the given text. To do so, we implement a variation of the Montecarlo particle filter to obtain a matching formation iteratively. In every iteration, we generate a set of new formations and evaluate their Clip Similarity with the given text, selecting the best formations according to this metric. This metric is obtained using Clip, [1], an existing foundation model trained to encode images and texts into vectors within a common latent space. The comparison between these vectors determines how likely the given text describes the shapes. Our initial proof of concept shows the potential of this solution to generate robot swarm formations just from natural language descriptions and demonstrates a novel application of foundation models, such as CLIP, in the field of multi-robot systems. In this first approach, we create formations using a Convex-Hull approach. Next steps include more robust and generic representation and optimization steps in the process of obtaining a suitable swarm formation.

In view of the classical visual servoing trajectory planning method which only considers the camera trajectory, this paper proposes one homography matrix based trajectory planning method for robot uncalibrated visual servoing. Taking the robot-end-effector frame as one generic case, eigenvalue decomposition is utilized to calculate the infinite homography matrix of the robot-end-effector trajectory, and then the image feature-point trajectories corresponding to the camera rotation is obtained, while the image feature-point trajectories corresponding to the camera translation is obtained by the homography matrix. According to the additional image corresponding to the robot-end-effector rotation, the relationship between the robot-end-effector rotation and the variation of the image feature-points is obtained, and then the expression of the image trajectories corresponding to the optimal robot-end-effector trajectories (the rotation trajectory of the minimum geodesic and the linear translation trajectory) are obtained. Finally, the optimal image trajectories of the uncalibrated visual servoing controller is modified to track the image trajectories. Simulation experiments show that, compared with the classical IBUVS method, the proposed trajectory planning method can obtain the shortest path of any frame and complete the robot visual servoing task with large initial pose deviation.

Different methods are used for a mobile robot to go to a specific target location. These methods work in different ways for online and offline scenarios. In the offline scenario, an environment map is created once, and fixed path planning is made on this map to reach the target. Path planning algorithms such as A* and RRT (Rapidly-Exploring Random Tree) are the examples of offline methods. The most obvious situation here is the need to re-plan the path for changing conditions of the loaded map. On the other hand, in the online scenario, the robot moves dynamically to a given target without using a map by using the perceived data coming from the sensors. Approaches such as SFM (Social Force Model) are used in online systems. However, these methods suffer from the requirement of a lot of dynamic sensing data. Thus, it can be said that the need for re-planning and mapping in offline systems and various system design requirements in online systems are the subjects that focus on autonomous mobile robot research. Recently, deep neural network powered Q-Learning methods are used as an emerging solution to the aforementioned problems in mobile robot navigation. In this study, machine learning algorithms with deep Q-Learning (DQN) and Deep DQN architectures, are evaluated for the solution of the problems presented above to realize path planning of an autonomous mobile robot to avoid obstacles.

Classical deterministic optimal control problems assume full information about the controlled process. The theory of control for general partially-observable processes is powerful, but the methods are computationally expensive and typically address the problems with stochastic dynamics and continuous (directly unobserved) stochastic perturbations. In this paper we focus on path planning problems which are in between -- deterministic, but with an initial uncertainty on either the target or the running cost on parts of the domain. That uncertainty is later removed at some time $T$, and the goal is to choose the optimal trajectory until then. We address this challenge for three different models of information acquisition: with fixed $T$, discretely distributed and exponentially distributed random $T$. We develop models and numerical methods suitable for multiple notions of optimality: based on the average-case performance, the worst-case performance, the average constrained by the worst, the average performance with probabilistic constraints on the bad outcomes, risk-sensitivity, and distributional-robustness. We illustrate our approach using examples of pursuing random targets identified at a (possibly random) later time $T$.

This paper considers a generalization of the Path Finding (PF) problem with refuelling constraints referred to as the Gas Station Problem (GSP). Similar to PF, given a graph where vertices are gas stations with known fuel prices, and edge costs are the gas consumption between the two vertices, GSP seeks a minimum-cost path from the start to the goal vertex for a robot with a limited gas tank and a limited number of refuelling stops. While GSP is polynomial-time solvable, it remains a challenge to quickly compute an optimal solution in practice since it requires simultaneously determine the path, where to make the stops, and the amount to refuel at each stop. This paper develops a heuristic search algorithm called Refuel A$^*$ (RF-A$^*$) that iteratively constructs partial solution paths from the start to the goal guided by a heuristic while leveraging dominance rules for pruning during planning. RF-A$^*$ is guaranteed to find an optimal solution and often runs 2 to 8 times faster than the existing approaches in large city maps with several hundreds of gas stations.

Relational networks within a team play a critical role in the performance of many real-world multi-robot systems. To successfully accomplish tasks that require cooperation and coordination, different agents (e.g., robots) necessitate different priorities based on their positioning within the team. Yet, many of the existing multi-robot cooperation algorithms regard agents as interchangeable and lack a mechanism to guide the type of cooperation strategy the agents should exhibit. To account for the team structure in cooperative tasks, we propose a novel algorithm that uses a relational network comprising inter-agent relationships to prioritize certain agents over others. Through appropriate design of the team's relational network, we can guide the cooperation strategy, resulting in the emergence of new behaviors that accomplish the specified task. We conducted six experiments in a multi-robot setting with a cooperative task. Our results demonstrate that the proposed method can effectively influence the type of solution that the algorithm converges to by specifying the relationships between the agents, making it a promising approach for tasks that require cooperation among agents with a specified team structure.

Multi-robot coordination has traditionally relied on a mission-specific and expert-driven pipeline, where natural language mission descriptions are manually translated by domain experts into mathematical formulation, algorithm design, and executable code. This conventional process is labor-intensive, inaccessible to non-experts, and inflexible to changes in mission requirements. Here, we propose LAN2CB (Language to Collective Behavior), a novel framework that leverages large language models (LLMs) to streamline and generalize the multi-robot coordination pipeline. LAN2CB transforms natural language (NL) mission descriptions into executable Python code for multi-robot systems through two core modules: (1) Mission Analysis, which parses mission descriptions into behavior trees, and (2) Code Generation, which leverages the behavior tree and a structured knowledge base to generate robot control code. We further introduce a dataset of natural language mission descriptions to support development and benchmarking. Experiments in both simulation and real-world environments demonstrate that LAN2CB enables robust and flexible multi-robot coordination from natural language, significantly reducing manual engineering effort and supporting broad generalization across diverse mission types. Website: https://sites.google.com/view/lan-cb

The superiority of Multi-Robot Systems (MRS) in various complex environments is unquestionable. However, in complex situations such as search and rescue, environmental monitoring, and automated production, robots are often required to work collaboratively without a central control unit. This necessitates an efficient and robust decentralized control mechanism to process local information and guide the robots' behavior. In this work, we propose a new decentralized controller design method that utilizes the Deep Q-Network (DQN) algorithm from deep reinforcement learning, aimed at improving the integration of local information and robustness of multi-robot systems. The designed controller allows each robot to make decisions independently based on its local observations while enhancing the overall system's collaborative efficiency and adaptability to dynamic environments through a shared learning mechanism. Through testing in simulated environments, we have demonstrated the effectiveness of this controller in improving task execution efficiency, strengthening system fault tolerance, and enhancing adaptability to the environment. Furthermore, we explored the impact of DQN parameter tuning on system performance, providing insights for further optimization of the controller design. Our research not only showcases the potential application of the DQN algorithm in the decentralized control of multi-robot systems but also offers a new perspective on how to enhance the overall performance and robustness of the system through the integration of local information.

We present a method for deadlock-free and collision-free navigation in a multi-robot system with nonholonomic robots. The problem is solved by quadratic programming and is applicable to most wheeled mobile robots with linear kinematic constraints. We introduce masked velocity and Masked Cooperative Collision Avoidance (MCCA) algorithm to encourage a fully decentralized deadlock avoidance behavior. To verify the method, we provide a detailed implementation and introduce heading oscillation avoidance for differential-drive robots. To the best of our knowledge, it is the first method to give very promising and stable results for deadlock avoidance even in situations with a large number of robots and narrow passages.

Rendezvous aims at gathering all robots at a specific location, which is an important collaborative behavior for multi-robot systems. However, in an unknown environment, it is challenging to achieve rendezvous. Previous researches mainly focus on special scenarios where communication is not allowed and each robot executes a random searching strategy, which is highly time-consuming, especially in large-scale environments. In this work, we focus on rendezvous in unknown environments where communication is available. We divide this task into two steps: rendezvous based environment exploration with relative pose (RP) estimation and rendezvous point selection. A new strategy called partitioned and incomplete exploration for rendezvous (PIER) is proposed to efficiently explore the unknown environment, where lightweight topological maps are constructed and shared among robots for RP estimation with very few communications. Then, a rendezvous point selection algorithm based on the merged topological map is proposed for efficient rendezvous for multi-robot systems. The effectiveness of the proposed methods is validated in both simulations and real-world experiments.

In this paper, we address the problem of real-time motion planning for multiple robotic manipulators that operate in close proximity. We build upon the concept of dynamic fabrics and extend them to multi-robot systems, referred to as Multi-Robot Dynamic Fabrics (MRDF). This geometric method enables a very high planning frequency for high-dimensional systems at the expense of being reactive and prone to deadlocks. To detect and resolve deadlocks, we propose Rollout Fabrics where MRDF are forward simulated in a decentralized manner. We validate the methods in simulated close-proximity pick-and-place scenarios with multiple manipulators, showing high success rates and real-time performance.

Multi-robot systems have become very popular in recent years because of their wide spectrum of applications, ranging from surveillance to cooperative payload transportation. Model Predictive Control (MPC) is a promising controller for multi-robot control because of its preview capability and ability to handle constraints easily. The performance of the MPC widely depends on many parameters, among which the prediction horizon is the major contributor. Increasing the prediction horizon beyond a limit drastically increases the computation cost. Tuning the value of the prediction horizon can be very time-consuming, and the tuning process must be repeated for every task. Moreover, instead of using a fixed horizon for an entire task, a better balance between performance and computation cost can be established if different prediction horizons can be employed for every robot at each time step. Further, for such variable prediction horizon MPC for multiple robots, on-demand collision avoidance is the key requirement. We propose Versatile On-demand Collision Avoidance (VODCA) strategy to comply with the variable horizon model predictive control. We also present a framework for learning the prediction horizon for the multi-robot system as a function of the states of the robots using the Soft Actor-Critic (SAC) RL algorithm. The results are illustrated and validated numerically for different multi-robot tasks.

A core challenge of multi-robot interactions is collision avoidance among robots with potentially conflicting objectives. We propose a game-theoretic method for collision avoidance based on rotating hyperplane constraints. These constraints ensure collision avoidance by defining separating hyperplanes that rotate around a keep-out zone centered on certain robots. Since it is challenging to select the parameters that define a hyperplane without introducing infeasibilities, we propose to learn them from an expert trajectory i.e., one collected by recording human operators. To do so, we solve for the parameters whose corresponding equilibrium trajectory best matches the expert trajectory. We validate our method by learning hyperplane parameters from noisy expert trajectories and demonstrate the generalizability of the learned parameters to scenarios with more robots and previously unseen initial conditions.

In this paper we propose a periodic solution to the problem of persistently covering a finite set of interest points with a group of autonomous mobile agents. These agents visit periodically the points and spend some time carrying out the coverage task, which we call coverage time. Since this periodic persistent coverage problem is NP-hard, we split it into three subproblems to counteract its complexity. In the first place, we plan individual closed paths for the agents to cover all the points. Second, we formulate a quadratically constrained linear program to find the optimal coverage times and actions that satisfy the coverage objective. Finally, we join together the individual plans of the agents in a periodic team plan by obtaining a schedule that guarantees collision avoidance. To this end, we solve a mixed integer linear program that minimizes the time in which two or more agents move at the same time. Eventually, we apply the proposed solution to an induction hob with mobile inductors for a domestic heating application and show its performance with experiments on a real prototype.

Trajectory replanning is a critical problem for multi-robot teams navigating dynamic environments. We present RLSS (Replanning using Linear Spatial Separations): a real-time trajectory replanning algorithm for cooperative multi-robot teams that uses linear spatial separations to enforce safety. Our algorithm handles the dynamic limits of the robots explicitly, is completely distributed, and is robust to environment changes, robot failures, and trajectory tracking errors. It requires no communication between robots and relies instead on local relative measurements only. We demonstrate that the algorithm works in real-time both in simulations and in experiments using physical robots. We compare our algorithm to a state-of-the-art online trajectory generation algorithm based on model predictive control, and show that our algorithm results in significantly fewer collisions in highly constrained environments, and effectively avoids deadlocks.

The safe and efficient operation of Autonomous Mobile Robots (AMRs) in complex environments, such as manufacturing, logistics, and agriculture, necessitates accurate multi-object tracking and predictive collision avoidance. This paper presents algorithms and techniques for addressing these challenges using Lidar sensor data, emphasizing ensemble Kalman filter. The developed predictive collision avoidance algorithm employs the data provided by lidar sensors to track multiple objects and predict their velocities and future positions, enabling the AMR to navigate safely and effectively. A modification to the dynamic windowing approach is introduced to enhance the performance of the collision avoidance system. The overall system architecture encompasses object detection, multi-object tracking, and predictive collision avoidance control. The experimental results, obtained from both simulation and real-world data, demonstrate the effectiveness of the proposed methods in various scenarios, which lays the foundation for future research on global planners, other controllers, and the integration of additional sensors. This thesis contributes to the ongoing development of safe and efficient autonomous systems in complex and dynamic environments.

We study the problem that requires a team of robots to perform joint localization and target tracking task while ensuring team connectivity and collision avoidance. The problem can be formalized as a nonlinear, non-convex optimization program, which is typically hard to solve. To this end, we design a two-staged approach that utilizes a greedy algorithm to optimize the joint localization and target tracking performance and applies control barrier functions to ensure safety constraints, i.e., maintaining connectivity of the robot team and preventing inter-robot collisions. Simulated Gazebo experiments verify the effectiveness of the proposed approach. We further compare our greedy algorithm to a non-linear optimization solver and a random algorithm, in terms of the joint localization and tracking quality as well as the computation time. The results demonstrate that our greedy algorithm achieves high task quality and runs efficiently.

Recent advances in multi-agent reinforcement learning (MARL) are enabling impressive coordination in heterogeneous multi-robot teams. However, existing approaches often overlook the challenge of generalizing learned policies to teams of new compositions, sizes, and robots. While such generalization might not be important in teams of virtual agents that can retrain policies on-demand, it is pivotal in multi-robot systems that are deployed in the real-world and must readily adapt to inevitable changes. As such, multi-robot policies must remain robust to team changes -- an ability we call adaptive teaming. In this work, we investigate if awareness and communication of robot capabilities can provide such generalization by conducting detailed experiments involving an established multi-robot test bed. We demonstrate that shared decentralized policies, that enable robots to be both aware of and communicate their capabilities, can achieve adaptive teaming by implicitly capturing the fundamental relationship between collective capabilities and effective coordination. Videos of trained policies can be viewed at: https://sites.google.com/view/cap-comm

Developing a safe, stable, and efficient obstacle avoidance policy in crowded and narrow scenarios for multiple robots is challenging. Most existing studies either use centralized control or need communication with other robots. In this paper, we propose a novel logarithmic map-based deep reinforcement learning method for obstacle avoidance in complex and communication-free multi-robot scenarios. In particular, our method converts laser information into a logarithmic map. As a step toward improving training speed and generalization performance, our policies will be trained in two specially designed multi-robot scenarios. Compared to other methods, the logarithmic map can represent obstacles more accurately and improve the success rate of obstacle avoidance. We finally evaluate our approach under a variety of simulation and real-world scenarios. The results show that our method provides a more stable and effective navigation solution for robots in complex multi-robot scenarios and pedestrian scenarios. Videos are available at https://youtu.be/r0EsUXe6MZE.

Multi-agent reinforcement learning (MARL) has emerged as a promising solution for learning complex and scalable coordination behaviors in multi-robot systems. However, established MARL platforms (e.g., SMAC and MPE) lack robotics relevance and hardware deployment, leaving multi-robot learning researchers to develop bespoke environments and hardware testbeds dedicated to the development and evaluation of their individual contributions. The Multi-Agent RL Benchmark and Learning Environment for the Robotarium (MARBLER) is an exciting recent step in providing a standardized robotics-relevant platform for MARL, by bridging the Robotarium testbed with existing MARL software infrastructure. However, MARBLER lacks support for parallelization and GPU/TPU execution, making the platform prohibitively slow compared to modern MARL environments and hindering adoption. We contribute JaxRobotarium, a Jax-powered end-to-end simulation, learning, deployment, and benchmarking platform for the Robotarium. JaxRobotarium enables rapid training and deployment of multi-robot RL (MRRL) policies with realistic robot dynamics and safety constraints, supporting parallelization and hardware acceleration. Our generalizable learning interface integrates easily with SOTA MARL libraries (e.g., JaxMARL). In addition, JaxRobotarium includes eight standardized coordination scenarios, including four novel scenarios that bring established MARL benchmark tasks (e.g., RWARE and Level-Based Foraging) to a robotics setting. We demonstrate that JaxRobotarium retains high simulation fidelity while achieving dramatic speedups over baseline (20x in training and 150x in simulation), and provides an open-access sim-to-real evaluation pipeline through the Robotarium testbed, accelerating and democratizing access to multi-robot learning research and evaluation. Our code is available at https://github.com/GT-STAR-Lab/JaxRobotarium.

We consider the problem of designing distributed collision-avoidance multi-agent control in large-scale environments with potentially moving obstacles, where a large number of agents are required to maintain safety using only local information and reach their goals. This paper addresses the problem of collision avoidance, scalability, and generalizability by introducing graph control barrier functions (GCBFs) for distributed control. The newly introduced GCBF is based on the well-established CBF theory for safety guarantees but utilizes a graph structure for scalable and generalizable decentralized control. We use graph neural networks to learn both neural a GCBF certificate and distributed control. We also extend the framework from handling state-based models to directly taking point clouds from LiDAR for more practical robotics settings. We demonstrated the efficacy of GCBF in a variety of numerical experiments, where the number, density, and traveling distance of agents, as well as the number of unseen and uncontrolled obstacles increase. Empirical results show that GCBF outperforms leading methods such as MAPPO and multi-agent distributed CBF (MDCBF). Trained with only 16 agents, GCBF can achieve up to 3 times improvement of success rate (agents reach goals and never encountered in any collisions) on <500 agents, and still maintain more than 50% success rates for >1000 agents when other methods completely fail.

Decentralized multi-agent navigation under uncertainty is a complex task that arises in numerous robotic applications. It requires collision avoidance strategies that account for both kinematic constraints, sensing and action execution noise. In this paper, we propose a novel approach that integrates the Model Predictive Path Integral (MPPI) with a probabilistic adaptation of Optimal Reciprocal Collision Avoidance. Our method ensures safe and efficient multi-agent navigation by incorporating probabilistic safety constraints directly into the MPPI sampling process via a Second-Order Cone Programming formulation. This approach enables agents to operate independently using local noisy observations while maintaining safety guarantees. We validate our algorithm through extensive simulations with differential-drive robots and benchmark it against state-of-the-art methods, including ORCA-DD and B-UAVC. Results demonstrate that our approach outperforms them while achieving high success rates, even in densely populated environments. Additionally, validation in the Gazebo simulator confirms its practical applicability to robotic platforms. A source code is available at http://github.com/PathPlanning/MPPI-Collision-Avoidance.

The combination of a small unmanned ground vehicle (UGV) and a large unmanned carrier vehicle allows more flexibility in real applications such as rescue in dangerous scenarios. The autonomous recovery system, which is used to guide the small UGV back to the carrier vehicle, is an essential component to achieve a seamless combination of the two vehicles. This paper proposes a novel autonomous recovery framework with a low-cost monocular vision system to provide accurate positioning and attitude estimation of the UGV during navigation. First, we introduce a light-weight convolutional neural network called UGV-KPNet to detect the keypoints of the small UGV from the images captured by a monocular camera. UGV-KPNet is computationally efficient with a small number of parameters and provides pixel-level accurate keypoints detection results in real-time. Then, six degrees of freedom pose is estimated using the detected keypoints to obtain positioning and attitude information of the UGV. Besides, we are the first to create a large-scale real-world keypoints dataset of the UGV. The experimental results demonstrate that the proposed system achieves state-of-the-art performance in terms of both accuracy and speed on UGV keypoint detection, and can further boost the 6-DoF pose estimation for the UGV.

Unmanned aerial vehicles (UAVs) are capable of surveying expansive areas, but their operational range is constrained by limited battery capacity. The deployment of mobile recharging stations using unmanned ground vehicles (UGVs) significantly extends the endurance and effectiveness of UAVs. However, optimizing the routes of both UAVs and UGVs, known as the UAV-UGV cooperative routing problem, poses substantial challenges, particularly with respect to the selection of recharging locations. Here in this paper, we leverage reinforcement learning (RL) for the purpose of identifying optimal recharging locations while employing constraint programming to determine cooperative routes for the UAV and UGV. Our proposed framework is then benchmarked against a baseline solution that employs Genetic Algorithms (GA) to select rendezvous points. Our findings reveal that RL surpasses GA in terms of reducing overall mission time, minimizing UAV-UGV idle time, and mitigating energy consumption for both the UAV and UGV. These results underscore the efficacy of incorporating heuristics to assist RL, a method we refer to as heuristics-assisted RL, in generating high-quality solutions for intricate routing problems.

Unmanned Aerial Vehicles (UAVs), although adept at aerial surveillance, are often constrained by limited battery capacity. By refueling on slow-moving Unmanned Ground Vehicles (UGVs), their operational endurance can be significantly enhanced. This paper explores the computationally complex problem of cooperative UAV-UGV routing for vast area surveillance within the speed and fuel constraints, presenting a sequential multi-agent planning framework for achieving feasible and optimally satisfactory solutions. By considering the UAV fuel limits and utilizing a minimum set cover algorithm, we determine UGV refueling stops, which in turn facilitate UGV route planning at the first step and through a task allocation technique and energy constrained vehicle routing problem modeling with time windows (E-VRPTW) we achieve the UAV route at the second step of the framework. The effectiveness of our multi-agent strategy is demonstrated through the implementation on 30 different task scenarios across 3 different scales. This work offers significant insight into the collaborative advantages of UAV-UGV systems and introduces heuristic approaches to bypass computational challenges and swiftly reach high-quality solutions.

Fast moving unmanned aerial vehicles (UAVs) are well suited for aerial surveillance, but are limited by their battery capacity. To increase their endurance UAVs can be refueled on slow moving unmanned ground vehicles (UGVs). The cooperative routing of UAV-UGV multi-agent system to survey vast regions within their speed and fuel constraints is a computationally challenging problem, but can be simplified with heuristics. Here we present multiple heuristics to enable feasible and sufficiently optimal solutions to the problem. Using the UAV fuel limits and the minimum set cover algorithm, the UGV refueling stops are determined. These refueling stops enable the allocation of mission points to the UAV and UGV. A standard traveling salesman formulation and a vehicle routing formulation with time windows, dropped visits, and capacity constraints is used to solve for the UGV and UAV route, respectively. Experimental validation on a small-scale testbed (http://tiny.cc/8or8vz) underscores the effectiveness of our multi-agent approach.

Efficient mission planning for cooperative systems involving Unmanned Aerial Vehicles (UAVs) and Unmanned Ground Vehicles (UGVs) requires addressing energy constraints, scalability, and coordination challenges between agents. UAVs excel in rapidly covering large areas but are constrained by limited battery life, while UGVs, with their extended operational range and capability to serve as mobile recharging stations, are hindered by slower speeds. This heterogeneity makes coordination between UAVs and UGVs critical for achieving optimal mission outcomes. In this work, we propose a scalable deep reinforcement learning (DRL) framework to address the energy-constrained cooperative routing problem for multi-agent UAV-UGV teams, aiming to visit a set of task points in minimal time with UAVs relying on UGVs for recharging during the mission. The framework incorporates sortie-wise agent switching to efficiently manage multiple agents, by allocating task points and coordinating actions. Using an encoder-decoder transformer architecture, it optimizes routes and recharging rendezvous for the UAV-UGV team in the task scenario. Extensive computational experiments demonstrate the framework's superior performance over heuristic methods and a DRL baseline, delivering significant improvements in solution quality and runtime efficiency across diverse scenarios. Generalization studies validate its robustness, while dynamic scenario highlights its adaptability to real-time changes with a case study. This work advances UAV-UGV cooperative routing by providing a scalable, efficient, and robust solution for multi-agent mission planning.

At Levels 2 and 3 of autonomous driving defined by the Society of Auto-motive Engineers, drivers must take on certain driving responsibilities, and automated driving must sometimes yield to human control. This situation can occur in real time due to uncertainties in sensor measurements caused by environmental factors like fog or smoke. To address this challenge, we propose a method to manage real-time sensor uncertainties in autonomous vehicles by monitoring sensor conflicts and dynamically adjusting control authority to maintain safe operation. However, to achieve this, we have introduced a novel metric called the Degree of Conflicts (DoC), which quantifies the conflict between real-time sensor data by measuring the differences between data from multiple sensors. Our approach aims to demonstrate the importance of selecting an appropriate DoC threshold for transferring control between the automation agent and the human driver. The results have shown that choosing the correct DoC threshold can enhance safety by promptly handing over the driving control from the automation system to the human driver in challenging conditions.

Real-time object detection takes an essential part in the decision-making process of numerous real-world applications, including collision avoidance and path planning in autonomous driving systems. This paper presents a novel real-time streaming perception method named CorrDiff, designed to tackle the challenge of delays in real-time detection systems. The main contribution of CorrDiff lies in its adaptive delay-aware detector, which is able to utilize runtime-estimated temporal cues to predict objects' locations for multiple future frames, and selectively produce predictions that matches real-world time, effectively compensating for any communication and computational delays. The proposed model outperforms current state-of-the-art methods by leveraging motion estimation and feature enhancement, both for 1) single-frame detection for the current frame or the next frame, in terms of the metric mAP, and 2) the prediction for (multiple) future frame(s), in terms of the metric sAP (The sAP metric is to evaluate object detection algorithms in streaming scenarios, factoring in both latency and accuracy). It demonstrates robust performance across a range of devices, from powerful Tesla V100 to modest RTX 2080Ti, achieving the highest level of perceptual accuracy on all platforms. Unlike most state-of-the-art methods that struggle to complete computation within a single frame on less powerful devices, CorrDiff meets the stringent real-time processing requirements on all kinds of devices. The experimental results emphasize the system's adaptability and its potential to significantly improve the safety and reliability for many real-world systems, such as autonomous driving. Our code is completely open-sourced and is available at https://anonymous.4open.science/r/CorrDiff.

Path planning is crucial for the navigation of autonomous vehicles, yet these vehicles face challenges in complex and real-world environments. Although a global view may be provided, it is often outdated, necessitating the reliance of Unmanned Ground Vehicles (UGVs) on real-time local information. This reliance on partial information, without considering the global context, can lead to UGVs getting stuck in local minima. This paper develops a method to proactively predict local minima using Dynamic Bayesian filtering, based on the detected obstacles in the local view and the global goal. This approach aims to enhance the autonomous navigation of self-driving vehicles by allowing them to predict potential pitfalls before they get stuck, and either ask for help from a human, or re-plan an alternate trajectory.