航空航天背景下的多星任务规划中的应急规划/突发任务规划

Emergency observation tasks involve Earth imaging activities by satellites to aid in emergency scenarios like disaster relief, emphasizing promptness and unpredictability. Multi-satellite collaborative planning for several emergency tasks faces issues like high computational complexity, slow processing, and numerous optimization considerations. This paper introduces a hierarchical planning approach and designs an algorithm for collaborative task planning across multiple emergencies and satellites. The algorithm’s computational efficiency and planning effectiveness are later confirmed through a case study.

Multi-satellite imaging mission planning (MSIMP) has been difficult in various scenarios due to the complex constraints of multi-satellite imaging, the wide area covered by target points, and the difficulty of achieving different mission requirements in a short period of time with limited satellite resources. In addressing this challenge, this work investigates multi-satellite imaging mission planning based on the Unified Plan Model and Improved Adaptive Differential Evolution algorithm (UPM-IADE). First, a unified model is built based on two scenarios: a large-scale imaging mission and an emergency support mission, and then a mission assignment framework is adaptively selected based on mission priority. Second, a monorail task synthesis method based on visible time windows is created to clarify the execution relationship between the satellite and the target point. Finally, an individual weight ranking rule is developed, and the weight is used to combine the fitness value ranking and diversity ranking into a final fitness value ranking, which is used to select individuals that satisfy the mutation requirements into the mutation strategy pool for adaptive mutation strategy selection. Experiments 1, 2, 3, and 4 have demonstrated that UPM-IADE can successfully resolve the imaging satellite mission planning for both scenarios while providing remarkable performance in terms of high mission benefit and rapid response.

Continuous observation is a problem that must be considered in the process of emergency observation of disasters. In essence, the continuous observation task can be understood as a multi-stage observation task, that is, repeated observation of a certain region according to a certain frequency. With the change of time, the actual situation of disaster area changes accordingly. Only by continuously acquiring remote sensing images of the target area, relevant departments can master the first-hand data of disaster situation changes within the region, and take the comparisons as a powerful basis to command and adjust the rescue and relief work. Based on the full understanding of the observation characteristics and coverage capacity of satellite and other kinds of observation resources, the sustainable collaborative planning method of heterogeneous resources for disaster emergency observation is explored. The simulation results show that the method is effective.

As the number of satellites and satellite tasks continues to increase, multi-satellite collaboration task planning faces several challenges, including high complexity, high timeliness and limited scalability. This paper presents a method for task planning based on behavior tree, which leverages modularity and hierarchical structure of behavior tree to simplify planning process and enhance timeliness. We also propose an intelligent planning algorithm combining variable neighborhood search (VNS) algorithm with backtracking search algorithm (BSA) to reduce solution space and improve convergence speed. Experimental results show that this method improves satellite task planning timeliness by 16.7%, overcomes the flexibility and timeliness disadvantages of traditional methods in complex environments, and highlights the significant advantages and potential applications of behavior tree in multi-satellite collaboration field.

This paper proposes a task planning method that integrates deep Q-learning network (DQN) with matrix sorting for Earth-oriented static multi-target cooperative observation tasks. The approach addresses emergent satellite failures in imaging constellations by eliminating the need for network model retraining during satellite malfunctions. It enables real-time generation of optimal task allocation schemes in contingency scenarios, ensuring efficient and adaptive task planning. Firstly, a mission scenario model is established by formulating task constraints and defining optimization objectives; secondly, a deep reinforcement learning framework is constructed to output the observation target sequence; then, the observation target sequence is transformed into a target sequence matrix, and a matrix-sorting planning method is proposed to carry out the optimal assignment of the task; lastly, a replanning method is designed for sudden satellite failure and insertion of urgent tasks. The experimental results show that the method has fast task planning speed, high task completion, and immediate task replanning capability.

ABSTRACT Satellite remote sensing plays a crucial role in emergency disaster monitoring. However, variations in orbital parameters and inconsistent coverage frequently lead to inefficient observation and underutilization of valuable satellite resources. Therefore, we propose a coordination framework based on deep reinforcement learning to optimize multi-satellite observation scheduling, enabling more rapid and accurate disaster sensing. Specifically, the approach consists of three key components. First, it involves compiling an on-orbit satellite resource database by aggregating various orbital parameters. Second, it refines satellite imaging modes based on the geometric principles of on-orbit observation. Third, it introduces a Multi-Satellite Planning algorithm based on Proximal Policy Optimization (PPO) deep reinforcement learning, which dynamically schedules off-nadir angles and imaging durations to achieve optimal observation configurations. Moreover, experiments conducted with historical satellite imagery demonstrate that the proposed method significantly outperforms comparable algorithms in terms of coordination efficiency and timeliness of disaster information acquisition. The results validate the feasibility of this theoretical approach for multi-satellite emergency sensing. Importantly, the method offers considerable value for enhancing satellite-based disaster response, command and dispatch efficiency, and life-saving operations.

No abstract available

In this article, we propose a solution to multi-satellite intelligent task planning using the multi-agent reinforcement learning (MARL) method. Fristly, we have developed a multi-satellite task planning model based on the Markov game framework. Furthermore, we have computationally designed a satellite state transition function to address the task planning problem and successfully solved it using the multi-agent proximal policy optimization (MAPPO) algorithm. Our experimental results demonstrate that the MARL method exhibits remarkable convergence speed and performance, delivering significant rewards in multi-scale task planning scenarios. Consequently, it proves to be a highly suitable approach for multi-satellite intelligent task planning.

Earth Observation Satellites (EOS) are critical for acquiring space-based information, supporting diverse applications from environmental monitoring to urban planning. The increasing demand for satellite imaging services necessitates efficient planning to handle complex and heterogeneous observation tasks. Traditional mission planning approaches, often relying on “single objective, single model, single algorithm” paradigms with meta-tasks and time windows, struggle to integrate these diverse requirements and face scalability issues with increasing task complexity. To address these limitations, this paper proposes a novel Grid Heat-driven Imaging Satellite Complex Task Planning Method. We introduce a Grid Heat-based Requirements Observation Model (GHROM) that unifies point, area, and moving targets into a single, time-variant geospatial heatmap, where grid cell “heat” represents observation priority. For this gridbased representation, we design a Satellite Discrete State Task Planning Model that transforms the problem into finding the highest heat gain path within a Directed Acyclic Graph (DAG) of discrete satellite states, where attitude transition constraints are embedded as graph edges. Based on this model, we develop a Grid Heat-Driven Dynamic Programming Algorithm (GMDPA). Extensive simulation experiments demonstrate the effectiveness and significant advantages of GMDPA, particularly in largescale scenarios, showcasing its superior performance in terms of solution quality and computational efficiency compared to traditional heuristic, metaheuristic, and genetic algorithms.

Complex heterogeneous scenarios with multiple mission requirement relationships, poor model scalability, resource conflicts during mission planning are the serious challenges currently facing the field of multi-satellite imaging mission planning (MSIMP). To solve this difficult problem, this paper proposes an Objective task-matching strategy and Improved adaptive differential evolution algorithm (OTMS-IADE). Firstly, the target task matching strategy for MSIMP in complex heterogeneous scenarios is constructed for multi-user, multi-satellite and multi-task situations, which overcomes the problem of poor scalability of the planning model in complex heterogeneous scenarios, and reduces the loss of resources caused by inappropriate task allocation; Secondly, to address the problem of low execution efficiency and long planning time due to large MSIMP solution space and complex constraints in complex heterogeneous scenarios, an improved adaptive differential evolution algorithm is proposed to reasonably trade-off the spatial search performance and the spatial exploitation performance to enhance the algorithm solution efficiency. Simulation experiments show that the OTMS-IADE algorithm for processing complex heterogeneous scenarios MSIMP has obvious advantages regarding task importance optimization and timeliness.

As the number and variety of remote sensing satellites continue to grow, user demands are becoming increasingly complex and diverse. Concurrently, there is an escalating requirement for timeliness in satellite observations, thereby augmenting the complexity of task processing and resource allocation. In response to these challenges, this paper proposes an innovative method for dynamic task planning in multi-source remote sensing satellite cooperative observations tailored to complex scenarios. In the task processing phase, this study develops a preprocessing model suitable for various types of targets, enabling the decomposition of complex scenes into multiple point targets for independent satellite observation, thereby reducing the complexity of the problem. In the resource allocation phase, a dynamic task planning algorithm for multi-satellite cooperative observation is designed to achieve dynamic and optimized scheduling of the processed point targets, catering to the needs of multi-source remote sensing satellites. Empirical validation demonstrated that this method effectively implements dynamic adjustment plans for point targets, comprehensively optimizing the number of observation targets, computation time, task priority, and satellite resource utilization, significantly enhancing the dynamic observation efficiency of remote sensing satellites.

Autonomous task planning plays a vital role in the application of the earth observation satellites constellation as it can make full use of satellite resources and increase task revenue. In this paper, a multi-satellite cooperative task planning architecture based on the distributed multi-agent system is established. And an improved multi-satellite cooperative task planning algorithm (MSCTP-CEM) based on co-evolution mechanism is designed. Finally, the advantages of the proposed architecture and MSCTP-CEM algorithm are verified through multiple simulation experiments.

No abstract available

No abstract available

The existing multi-satellite dynamic mission planning system hardly satisfies the requirements of fast response time and high mission benefit in highly dynamic situations. In the meantime, a reasonable decision-maker preference mechanism is an additional challenge for multi-satellite imaging dynamic mission planning based on user preferences (MSDMPUP). Therefore, this study proposes the hybrid preference interaction mechanism and knowledge transfer strategy for the multi-objective evolutionary algorithm (HPIM–KTSMOEA). Firstly, an MSDMPUP model based on a task rolling window is constructed to achieve timely updating of the target task importance degree through the simultaneous application of periodic triggering and event triggering methods. Secondly, the hybrid preference interaction mechanism is constructed to plan according to the satellite controller’s preference-based commands in different phases of the optimal search of the mission planning scheme to effectively respond to the dynamic changes in the environment. Finally, a knowledge transfer strategy for the multi-objective evolutionary algorithm is proposed to accelerate population convergence in new environments based on knowledge transfer according to environmental variability. Simulation experiments verify the effectiveness and stability of the method in processing MSDMPUP. This study found that the HPIM–KTSMOEA algorithm has high task benefit, short response time, and high task completion when processing MSDMPUP.

Existing multi-satellite imaging mission planning methods mostly apply to scenarios with clear problem boundaries, explicit decision rules, and limited problem sizes. It is difficult to deal with the problem that the real environment is more complex and the action effect is more uncertain. This paper proposes a Human-machine interactive mechanism based on knowledge transfer strategy multi-objective evolutionary algorithm (HMIM-KTSMOEA) for Multi-satellite Imaging Mission Planning in Complex Environments (MSIMPCE). Firstly, a Human-machine interactive mechanism is proposed to deal with the dynamic changes of the environment effectively. Secondly, a multi-objective evolutionary algorithm based on a knowledge transfer strategy is proposed to improve the efficiency of MSIMPCE, which has complex constraints and large solving space. Simulation results show that HMIM-KTSMOEA has obvious advantages in solving the MSIMPCE problem in terms of task execution efficiency and response time.

Aiming at the imaging problem of regional target in emergency surveying and mapping, a multi-objective optimization model(MOOM) is proposed, which takes the imaging lateral swing angles of satellite as decision variables and takes the maximum coverage rate of regional target and the minimum number of holes as objective functions. Aiming at the two key problems of evaluation function calculation and multi-objective model solving, Vatti algorithm and NSGAII algorithm are used to solve them respectively. Finally, STK simulation data are used to verify the feasibility of the optimization method.

Wildfires spread quickly and require frequent satellite observations for early detection and effective management. Large Earth observation satellite constellations (EOSC) can meet this need with their broad coverage and diverse spectral capabilities. However, designing an efficient scheduling strategy for revisit tasks remains challenging due to the complex time‐coupled requirements and the multi‐objective nature of large‐scale EOSC operations. To address this, we introduce a time‐driven multi‐objective (TDMO) scheduling method. The key innovation of TDMO lies in its explicit integration of revisit intervals and a time‐driven mechanism, ensuring consistent observation frequencies and improved coordination of resources. Experiments across different wildfire scenarios show that TDMO effectively enhances scheduling efficiency and monitoring performance, offering a novel solution for dynamic and complex revisit scheduling in wildfire management.

To address the task scheduling problem of wide-ra nge high-resolution remote sensing satellite constellations for eme rgency disaster reduction, a spatio-temporal demand matrix for r emote sensing tasks was proposed.A remote sensing task scheduli ng algorithm based on moving candidate strips was designed on t his basis. Through simulation, the spatio-temporal demand matri x can flexibly describe the spatio-temporal requirements of differ ent remote sensing tasks. The remote sensing observation scheme can be generated through the adaptability of the scheduling algor ithm, which can basically meet the requirements of emergency dis aster reduction tasks.

This article investigates the satellite constellation scheduling problem for noncooperative space-object tracking. Different from the traditional Earth observation tasks, space-object tracking tasks require observation of fast moving targets that may have unpredictable trajectories, and estimation of the position of the targets through cooperative observation of multiple satellites. An agent-based scheduling framework is proposed to enable real-time, distributed scheduling of low Earth orbits constellation satellites for tracking noncooperative space targets. In the proposed framework, a nonlinear filtering of the targets’ trajectories is incorporated, and the observation satellites are dynamically allocated according to the status of the satellites and the states of the noncooperative targets. Specifically, event triggers of tracking space targets for activating task handover among satellites are defined and classified into five categories. The event-trigger processing and task response mechanism are embedded in the proposed scheduling framework. Thereafter, autonomously adjusting the observation-satellite sets can be achieved through distributed negotiation within only a few satellites. The effectiveness of the proposed method is validated by simulating scenarios of tracking targets without or with maneuvers.

No abstract available

Deploying multi-satellite constellations for Earth observation requires coordinating potentially hundreds of spacecraft. With increasing on-board capability for autonomy, we can view the constellation as a multi-agent system (MAS) and employ decentralized scheduling solutions. We formulate the problem as a distributed constraint optimization problem (DCOP) and desire scalable inter-agent communication. The problem consists of millions of variables which, coupled with the structure, make existing DCOP algorithms inadequate for this application. We develop a scheduling approach that employs a well-coordinated heuristic, referred to as the Geometric Neighborhood Decomposition (GND) heuristic, to decompose the global DCOP into sub-problems as to enable the application of DCOP algorithms. We present the Neighborhood Stochastic Search (NSS) algorithm, a decentralized algorithm to effectively solve the multi-satellite constellation observation scheduling problem using decomposition. In full, we identify the roadblocks of deploying DCOP solvers to a large-scale, real-world problem, propose a decomposition-based scheduling approach that is effective at tackling large scale DCOPs, empirically evaluate the approach against other baseline algorithms to demonstrate the effectiveness, and discuss the generality of the approach.

With the increasing demand for observations and the development of satellite technology, the scale of the Earth Observation Satellites (EOS) constellation is growing. To make the constellation more efficient during operation, scheduling imaging tasks of large-scale constellations is imperative. To solve the scheduling problem, we designed an allocation strategy for the daily operation and a Deep reinforcement learning (DRL) method for in-orbit emergencies. The results show that both of the method is effective.

No abstract available

The size and capabilities of Earth-observing satellite constellations are rapidly increasing. Leveraging distributed onboard control, we can enable novel time-sensitive measurements and responses. However, deploying autonomy to large multiagent satellite systems necessitates algorithms with efficient computation and communication. We tackle this challenge and propose new, online algorithms for large-scale dynamic distributed constraint optimization problems (DDCOP). We present the Dynamic Multi-Satellite Constellation Observation Scheduling Problem (DCOSP), a new formulation of DDCOPs that models integrated scheduling and execution. We construct an omniscient offline algorithm to compute the novel optimality condition of DCOSP and present the Dynamic Incremental Neighborhood Stochastic Search (D-NSS) algorithm, an incomplete online decomposition-based DDCOP approach. We show through simulation that D-NSS converges to near-optimal solutions and outperforms DDCOP baselines in terms of solution quality, computation time, and message volume. Our work forms the foundation of the largest in-space demonstration of distributed multiagent AI to date: the NASA FAME mission.

With the increasing communication and remote sensing demand, a constellation composed of multiple low-orbit satellites is usually designed. However, it is faced with the problems of how to achieve coverage with high frequency, long duration, and fully autonomous for a sudden ground target. Given the difficulty of the conventional constellation to achieve the above capabilities at a lower cost, a broken-chain constellation is proposed. The broken-chain constellation maintains a chain-like configuration in space to achieve better coverage uniformity, which can reduce the number of satellites. Aiming at the fully autonomous operation of the constellation, a distributed autonomous scheduling strategy is designed considering the chain-like characteristics. Compared with the traditional centralized scheduling method, this strategy not only effectively reduces the computational cost but also improves the dynamic response to the observation of sudden ground targets.

Commercial companies have launched hundreds of small, low-cost nanosatellites for Earth observation tasks, such as monitoring forest fires, landslides, and lake algae. These time-critical events require low-latency detection. Previous work has shown that capture latency, defined as the duration between an event occurrence and a satellite capturing an image covering the event region, is the major latency component. In this work, we ask what the minimum capture latency is under physical orbital dynamics and energy constraints. We first characterize how region location and satellite orbital configuration affect the capture latency. Next, we explore the scheduling of satellite captures when energy constraints prevent the capture of all images. We introduce Geoduck, a satellite capture scheduler specifically designed to minimize capture latency. Geoduck leverages the insight that the timing of many events follows a prior distribution; for example, forest fires are more likely to occur in summer. We analyze the impact of different capture schedules on latency and propose a dynamic programming algorithm to choose the optimal captures. Our evaluation shows that Geoduck reduces capture latency by 1.5--2.1×.

With the expanding scope of human activities in marine environments, the efficient detection and tracking of mobile targets on the ocean's surface have become increasingly crucial. Synthetic aperture radar (SAR) constellation can obtain ground observation data based on user requests and subject to visibility conditions. Now it is an indispensable tool in sea surface moving target search tasks. Satellite constellation resources are scarce and limited, and user demands are diverse. How to rationally dispatch satellite constellation resources to meet user needs to the maximum extent and improve the application efficiency of satellite resources is an urgent scientific problem that needs to be solved. This article mainly expounds two respects of work. First, modeling SAR constellation scheduling problem for sea surface moving target search tasks to establish the objective function. Second, a novel multistrategy discrete constrained differential evolution algorithm denoted as MSDCDE is proposed in the article. The proposed MSDCDE algorithm integrates cross strategy based on discrete variables, constraint handling techniques, population restart strategy, and left-shift local strategy, which can effectively avoid falling into local optimality, thereby achieving global optimality and improving search and rescue performances. Six sets of experiments, totaling 215 runs, have been conducted to validate the effectiveness of the proposed resolution process framework and the MSDCDE algorithm. The proposed method demonstrated an over 48.98% performance improvement compared with some state-of-the-art algorithms and significantly reduced task completion time.

Earth observation satellites (EOSs) play a pivotal role in capturing and analyzing planetary phenomena, ranging from natural disasters to societal development. The EOS scheduling problem (EOSSP), which optimizes the schedule of EOSs, is often solved with respect to nadir-directional EOS systems, thus restricting the observation time of targets and, consequently, the effectiveness of each EOS. This paper leverages state-of-the-art constellation reconfigurability to develop the reconfigurable EOS scheduling problem (REOSSP), wherein EOSs are assumed to be maneuverable, forming a more optimal constellation configuration at multiple opportunities during a schedule. This paper develops a novel mixed-integer linear programming formulation for the REOSSP to optimally solve the scheduling problem for given parameters. Additionally, since the REOSSP can be computationally expensive for large-scale problems, a rolling horizon procedure (RHP) solution method is developed. The performance of the REOSSP is benchmarked against the EOSSP, which serves as a baseline, through a set of random instances where problem characteristics are varied and a case study in which Hurricane Sandy is used to demonstrate realistic performance. These experiments demonstrate the value of constellation reconfigurability in its application to the EOSSP, yielding solutions that improve performance, while the RHP enhances computational runtime for large-scale REOSSP instances.

Emergency observations are missions executed by Earth observation satellites to support urgent ground operations. Emergency observations become more important for meeting the requirements of highly dynamic and highly time-sensitive observation missions, such as disaster monitoring and early warning. Considering the complex scheduling problem of Earth observation satellites under emergency conditions, a multi-satellite dynamic mission scheduling model based on mission priority is proposed in this paper. A calculation model of mission priority is designed for emergency missions based on seven impact factors. In the satellite mission scheduling, the resource constraints of scheduling are analyzed in detail, and the optimization objective function is built to maximize the observation mission priority and mission revenues, and minimize the waiting time for missions that require urgency for execution time. Then, the hybrid genetic tabu search algorithm is used to obtain the initial satellite scheduling plan. In case of the dynamic arrival of new emergency missions before scheduling plan releases, a dynamic scheduling algorithm based on mission priority is proposed to solve the scheduling problem caused by newly arrived missions and to obtain the scheduling plan of newly arrived missions. A simulation experiment was conducted for different numbers of initial missions and newly arrived missions, and the scheduling results were evaluated with a model performance evaluation function. The results show that the execution probability of high-priority missions increased because the mission priority was taken into account in the model. In the case of more satellite resources, when new missions dynamically arrived, the satellite resources can be reasonably allocated to these missions based on the mission priority. Overall, this approach reduces the complexity of the dynamic adjustment and maintains the stability of the initial scheduling plan.

Multi-satellite collaborative computing has achieved task decomposition and collaborative execution through inter-satellite links (ISLs), which has significantly improved the efficiency of task execution and system responsiveness. However, existing methods focus on single-task execution and lack multi-task parallel processing capability. Most methods ignore task priorities and dependencies, leading to excessive waiting times and poor scheduling results. To address these problems, this paper proposes a task decomposition and resource mapping method based on task priorities and resource constraints. First, we introduce a graph theoretic model to represent the task dependency and priority relationships explicitly, combined with a novel algorithm for task decomposition. Meanwhile, we construct a resource allocation model based on game theory and combine it with deep reinforcement learning to achieve resource mapping in a dynamic environment. Finally, we adopt the theory of temporal logic to formalize the execution order and time constraints of tasks and solve the dynamic scheduling problem through mixed-integer nonlinear programming to ensure the optimality and real-time updating of the scheduling scheme. The experimental results demonstrate that the proposed method improves resource utilization by up to about 24% and reduces overall execution time by up to about 42.6% in large-scale scenarios.

No abstract available

The purpose of constructing onboard observation mission queues is to improve the execution efficiency of onboard tasks and reduce energy consumption, representing a significant challenge in achieving efficient global military reconnaissance and target tracking. Existing research often focuses on the aspect of task scheduling, aiming at optimizing the efficiency of single-task execution, while neglecting the complex dependencies that might exist between multiple tasks and payloads. Moreover, traditional task scheduling schemes are no longer suitable for large-scale tasks. To effectively reduce the number of tasks within the network, we introduce a network aggregation graph model based on multiple satellites and tasks, and propose a task aggregation priority dynamic calculation algorithm based on graph computations. Subsequently, we present a dynamic merging-based method for multi-satellite, multi-task aggregation, a novel approach for constructing onboard mission queues that can dynamically optimize the task queue according to real-time task demands and resource status. Simulation experiments demonstrate that, compared to baseline algorithms, our proposed task aggregation method significantly reduces the task size by approximately 25% and effectively increases the utilization rate of onboard resources.

No abstract available

No abstract available

With the rapid increase in the number of Earth Observation Satellites (EOSs), research on autonomous mission scheduling has become increasingly critical for optimizing satellite sensor operations. While most existing studies focus on static environments or initial planning states, few address the challenge of dynamic request replanning for real-time sensor management. In this paper, we tackle the problem of multi-satellite rapid mission replanning under dynamic batch-arrival observation requests. The objective is to maximize overall observation revenue while minimizing disruptions to the original scheme. We propose a framework that integrates stochastic master-satellite mission allocation with single-satellite replanning, supported by reactive scheduling policies trained via deep reinforcement learning. Our approach leverages mission sequence modeling with attention mechanisms and time-attitude-aware rotary positional encoding to guide replanning. Additionally, scalable embeddings are employed to handle varying volumes of dynamic requests. The mission allocation phase efficiently generates assignment solutions using a pointer network, while the replanning phase introduces a hybrid action space for direct task insertion. Both phases are formulated as Markov Decision Processes (MDPs) and optimized using the PPO algorithm. Extensive simulations demonstrate that our method significantly outperforms state-of-the-art approaches, achieving a 15.27% higher request insertion revenue rate and a 3.05% improvement in overall mission revenue rate, while maintaining a 1.17% lower modification rate and achieving faster computational speeds. This demonstrates the effectiveness of our approach in real-world satellite sensor applications.

The exponential growth of Low Earth Orbit (LEO) satellites has revolutionised Earth Observation (EO) missions, addressing challenges in climate monitoring, disaster management, and more. However, autonomous coordination in multi-satellite systems remains a fundamental challenge. Traditional optimisation approaches struggle to handle the real-time decision-making demands of dynamic EO missions, necessitating the use of Reinforcement Learning (RL) and Multi-Agent Reinforcement Learning (MARL). In this paper, we investigate RL-based autonomous EO mission planning by modelling single-satellite operations and extending to multi-satellite constellations using MARL frameworks. We address key challenges, including energy and data storage limitations, uncertainties in satellite observations, and the complexities of decentralised coordination under partial observability. By leveraging a near-realistic satellite simulation environment, we evaluate the training stability and performance of state-of-the-art MARL algorithms, including PPO, IPPO, MAPPO, and HAPPO. Our results demonstrate that MARL can effectively balance imaging and resource management while addressing non-stationarity and reward interdependency in multi-satellite coordination. The insights gained from this study provide a foundation for autonomous satellite operations, offering practical guidelines for improving policy learning in decentralised EO missions.

With the rapid development of multi-satellite collaborative computing, most space tasks are interrelated with complex dependencies, often scheduled as DAG — the core scheduling paradigm for space-air-ground integrated applications. However, satellite network dynamics (node mobility) coupled with DAG task characteristics pose severe scheduling challenges. This study prioritizes reducing task completion latency while improving resource utilization, aiming to provide a reliable solution for complex space tasks in dynamic environments.Existing methods either rely on heuristics (lacking dynamic adaptability) or single reinforcement learning RL(ignoring DAG task dependency priorities, causing uneven resource allocation). To address this gap, this paper proposes an innovative algorithm integrating upward ranking for DAG task priority quantification with PPO. It captures task dependencies, dynamically adjusts strategies via RL, and better adapts to satellites’ limited computing resources for real-time dynamic response.Experimental results in a simulated multi-satellite network show the proposed algorithm reduces average execution latency by 30%-48%, with the gap widening as task nodes increase) and maintains resource utilization above 70% compared to traditional heuristics and single RL methods. This work provides an effective technical path for DAG task scheduling in dynamic satellite networks, advancing multi-satellite collaborative computing in remote sensing data processing and emergency communications.

No abstract available

No abstract available

The proliferation of satellite constellations, coupled with reduced tasking latency and diverse sensor capabilities, has expanded the opportunities for automated Earth observation. This paper introduces a fully automated Tip-and-Cue framework designed for satellite imaging tasking and scheduling. In this context, tips are generated from external data sources or analyses of prior satellite imagery, identifying spatiotemporal targets and prioritizing them for downstream planning. Corresponding cues are the imaging tasks formulated in response, which incorporate sensor constraints, timing requirements, and utility functions. The system autonomously generates candidate tasks, optimizes their scheduling across multiple satellites using continuous utility functions that reflect the expected value of each observation, and processes the resulting imagery using artificial-intelligence-based models, including object detectors and vision-language models. Structured visual reports are generated to support both interpretability and the identification of new insights for downstream tasking. The efficacy of the framework is demonstrated through a maritime vessel tracking scenario, utilizing Automatic Identification System (AIS) data for trajectory prediction, targeted observations, and the generation of actionable outputs. Maritime vessel tracking is a widely researched application, often used to benchmark novel approaches to satellite tasking, forecasting, and analysis. The system is extensible to broader applications such as smart-city monitoring and disaster response, where timely tasking and automated analysis are critical.

No abstract available

No abstract available

The Earth Observation planning community has access to tools that can propagate orbits and compute coverage of Earth observing imagers with customizable shapes and orientation, model the expected Earth Reflectance at various bands, epochs and directions, generate simplified instrument performance metrics for imagers and radars, and schedule single and multiple spacecraft payload operations. We are working toward integrating existing tools to design a planner that allows commercial small spacecraft to assess the opportunities for cross-calibration of their sensors against current satellite to be calibrated, specifications of the reference instruments, sensor stability, allowable latency between calibration measurements, differences in viewing and solar geometry between calibration measurements, etc. The planner would output cross-calibration opportunities for every reference target pair as a function of flexible user-defined parameters. We use a preliminary version of this planner to inform the design of a constellation of transfer radiometers that can serve as stable, radiometric references for commercial sensors to cross-calibrate with. We propose such a constellation for either vicarious cross-calibration using pre-selected sites, or top of the atmosphere (TOA) cross-calibration globally. Results from the calibration planner applied to a subset of informed architecture designs show that a 4 sat constellation provides multiple calibration opportunities within half a day planning horizon, for Cubesat sensors deployed into a typical rideshare orbits. While such opportunities are available for cross calibration image pairs within 5 deg of solar or view directions, and with-in an hour (for TOA) and less than a day (vicariously), the planner allows us to identify many more by relaxing user-defined restrictions.

Forest fire monitoring is crucial for ecological environmental protection and disaster prevention. Traditional methods of forest fire prevention and suppression often suffer from delayed responses and limited coverage. This makes it difficult to meet the demands of real-time monitoring. The development of remote sensing satellite technology offers a promising solution through multi-source collaborative observation. However, existing methods for calculating satellite revisit intervals have limitations in accuracy and adaptability. Multi-satellite collaborative observation can significantly enhance monitoring efficiency. To optimize satellite observation efficiency in forest fire monitoring. This paper proposes a new scheduling method based on multi-satellite collaboration. This method establishes a complementary monitoring system through coordinated multi-satellite observations. Three sets of comparative experiments were conducted in the southwestern forest region. These experiments systematically evaluated the monitoring effectiveness of different observation strategies. The proposed method addresses issues such as low resource utilization and poor timeliness in multi-satellite collaborative observation. It provides scientific decision-making support for forest fire emergency response. The experimental results show that the proposed method outperforms other approaches across all evaluation metrics. It significantly enhances the system’s response capability to sudden fire incidents. This method holds significant practical value for improving forest fire monitoring capabilities. It has the potential to play a crucial role in enhancing the overall effectiveness of forest fire management and emergency response.

Natural disasters are global issues that have led to huge casualties and property losses. Satellite observation is one of the predominant methods of disaster monitoring, and high-mobility satellites can normally provide faster responses. A study of the high-mobility satellite task planning problem for global decentralized multi-target observation is conducted in this paper. First, a “trajectory-target” database composed of candidate trajectories and the corresponding visibilities of targets is established. Second, a splicing algorithm is devised to connect the individual trajectories as a whole so that the task planning problem can be converted to a combinational optimization problem of candidate trajectories. To solve the optimization problem, a heuristic algorithm is designed based on NSGA-II with a double-layer encoding structure and hybrid-heuristic mutation operators. The results presented at the end of the paper verify the effectiveness of the proposed algorithms.

Much attention has been paid to how limited in-orbit satellite resources can be better utilized to meet the increasingly heavy demand for space observation. A variety of single-stage optimization problems have been discussed, with few considerations regarding the effect of the two stages in multi-satellite cooperative observation mission planning. In this study, bilevel programming is applied to simultaneously consider both mission assignment and satellite scheduling. The purpose of the bilevel programming model is to optimize the planning scheme of multi-satellite cooperative observation mission and maximize the comprehensive benefit from the perspective of the system as a whole. The upper level of the model formulates the mission assignment scheme, and the lower level determines the optimal resource scheduling scheme by a mathematical method on the basis of the upper level, and then feeds the results back to the upper level. The upper level and lower level affect each other, and the optimal solution is obtained through an iterative process under the solution framework of genetic algorithm. Extensive experiments are simulated to demonstrate the feasibility and efficiency of the proposed bilevel programming model.

No abstract available

Wildfire disasters have become increasingly frequent, leading to serious ecological disasters and causing a large number of casualties. Remote sensing plays a crucial role in wildfire emergency observation, but separate satellite observations usually cannot meet the requirements for timely and accurate monitoring. Collaborative multi‐satellite remote sensing is a feasible approach to fire emergency observation and has received extensive attention. However, developing an effective multi‐satellite observation plan is a complex NP‐hard problem due to the varying satellite orbit and sensor configurations. This paper presents the Immune Mechanism Non‐Dominated Sorting Genetic Algorithm II (IM‐NSGA‐II) to address this challenge. This study (1) aligns satellite sensor resources with wildfire characteristics through a dataset based on disaster monitoring needs; (2) designs a generation model for Satellite Observation Stripe Coverage Possibilities (SCP) to create the Set of SCP (SCP_Set) for each satellite as solution space for multi‐satellite observation scheduling; (3) proposes an optimization method of hybrid immune operation and NSGA‐II to determine the optimal multi‐satellite observation scheme for wildfire. An experiment is conducted on wildfires along the China‐Mongolia border to verify the effectiveness of the proposed algorithm in the field of wildfire emergency observation. This algorithm provides a novel approach for multi‐satellite collaborative observation in disaster response and other fields.

No abstract available

No abstract available

Space situation awareness (SSA) is a crucial technology for ensuring space security. With the increase in the number of space targets and the diversification of spacecraft sensing requirements, planning technology for multi-target sensing missions has become an indispensable part. This paper proposes a multi-satellite hierarchical collaborative mission planning method for multi-target sensing missions, aiming at the binocular or monocular orbit determination requirements of space targets. Firstly, a collaborative mission planning model for multi-target sensing missions is proposed to decouple multiple coupled complex constraints and design a hierarchical bidding and tendering collaboration framework. On this basis, a multi-satellite collaborative mission planning method is proposed, and a hybrid reward function related to binocular orbit determination observation configuration evaluation, attitude maneuver cost, and binocular collaborative reward is designed to guide multiple inter-satellite collaborations and ultimately obtain a multi-satellite collaborative observation sequence. Simulations verify the effectiveness and robustness of the proposed method through the Monte Carlo method, thereby providing theoretical and technical support for SSA applications.

There are various types of natural disasters, and the demand for reasonable scheduling of remote sensing satellites for natural disaster monitoring is very urgent. At this stage, there is no good method to match disasters and satellite resources. This paper studies the intelligent matching technology of natural disaster monitoring needs. This paper analyzes the key technologies of intelligent matching of disaster needs, mainly including combing the available satellite and payload resources, analyzing the types of natural disasters and satellite image requirements, and forming the satellite resource matching scheme through satellite resource modeling and preliminary transit analysis of satellite resources, so as to optimize the most suitable scheme. The intelligent matching scheme technology of satellite resources based on multi satellite and multi load and demand satisfaction is proposed. The intelligent matching technology can timely match appropriate observation resources according to different disaster needs, meet the multi satellite and multi load resource scheduling of various disaster tasks, and scientifically arrange satellite imaging. Fully improve the efficiency of satellite in orbit work for natural disaster monitoring.

With the burgeoning of remote sensing and space technology, multi-satellite collaborative mission planning, which is the key to achieving efficient Earth observation, has become increasingly intricate due to the expanding complexity and volume of observation missions. Addressing the multi-satellite collaborative mission planning problem, which is characterized by its two-stage decision-making process involving mission assignment and resource scheduling, this study investigates a comprehensive joint decision making that encompasses both mission assignment and resource scheduling and comprehensively optimizes the mission completion rate, the mission profit rate, and the satellite resource utilization rate. Considering the interaction of these decisions, we formulate the problem as a bilevel programming model from a game-theoretic perspective and propose a nested bilevel improved genetic algorithm (NBIGA) for its solution. Simulation experiments substantiate the applicability of the bilevel programming model in joint decision making for the stages of mission assignment and resource scheduling in multi-satellite collaborative mission planning, as well as the robustness of the NBIGA. A comparative analysis with the nested bilevel genetic algorithm (NBGA) confirms that the algorithm proposed in this study can achieve superior optimization outcomes and higher solving efficiency.

Space-based Earth observation is now developing rapidly because of the advantages in coverage, convenience, and flexibility. However, it is difficult for one single satellite to realize the quick observation of wide areas. In order to catch all the significant information of a wide area in a short time, multi-satellite observation mission would be proposed. In this article, a mission planning method for online multi-satellite wide area observation is established to serve future multi-satellite observation missions. Firstly, a method for area division is proposed, and the whole area is divided into subareas. Then the multi-satellite observation path planning is realized by a strategy of path deduction. After that, a remaining time allocation method to maximize the observation gazing time of each subarea is proposed. Finally, the algorithms for the whole mission planning process are provided. Numerical simulations show that the mission planning method is able to ensure the complete coverage of different wide target areas, with high reliability and low computational complexity.

A unified description for the imaging mission of different types of targets is lacking, making overall optimization of imaging missions of complex multitype targets (point, curve, and area) within a single pass difficult when using traditional satellite imaging mission planning. We propose an imaging mission planning modeling method based on the optimal mission decomposition/merge (MD/M) strategy for imaging missions of multitype targets within a single pass of SA-EOSs. This method transforms the imaging missions of multitype targets into an atomic mission set that can be described uniformly for integrated optimization scheduling. First, an optimal MD/M strategy was proposed for different types of target imaging (point, curve, and area) based on the characteristics of dynamic imaging. A mission optimization model was then constructed, with imaging coverage benefit and mission execution time as the objective functions, and an improved particle swarm optimization algorithm was used to solve the model. Finally, the proposed optimal MD/M strategy and mission planning modeling method were tested by setting up seven groups of imaging mission simulation experiments with different multi-type target combinations. The result showed that the proposed “two-stage” optimization method achieved integrated optimization of multimode imaging missions in dynamic imaging. The proposed optimal MD/M strategy can be applied in integrated modeling of imaging missions of multitype targets. Moreover, the imaging mission planning model constructed with time-stamped strips as atomic tasks can perform efficient integrated planning of imaging mission of complex multitype targets and ensures the effective performance of SA-EOSs.

Earth observation satellites, particularly agile Earth observation satellites (AEOSs) with enhanced attitude maneuverability, have become increasingly crucial in emergency response and disaster monitoring operations. Efficient mission planning for densely distributed ground targets with diverse priorities poses significant challenges, especially when considering strict attitude maneuver constraints and time-sensitive requirements. To address these challenges, this paper proposes a target clusters and dual-timeline optimization (TCDO) framework that integrates priority-based geographical clustering with temporal–spatial coordination mechanisms for efficient mission planning. The proposed approach effectively maintains satellite maneuver constraints while achieving significant improvements in priority-based target acquisition and computational efficiency. Experimental results demonstrate the framework’s superior performance, achieving a 94% coverage rate and a 99.5% reduction in computation time compared to traditional scheduling methods, such as linear programming and genetic algorithms.

Multi-satellite imaging mission planning (MSIMP) research has advanced substantially in recent years. However, contemporary MSIMP research in uncertain environments is still confronting challenges such as loss of satellite resource allocation, inadequate anti-jamming ability of the mission planning scheme, and low mission completion rate. Therefore, in this work, we propose a hybrid local replanning strategy improved adaptive differential evolutionary (HLRS-MSFADE) algorithm based on the multi-satellite imaging mission planning in uncertain environments (MSIMPUE). First, an MSIMPUE model based on uncertainty assessment is constructed. This model solves the problem of assessing new tasks with varied qualities to decide the observation order in an uncertain environment and decreases the loss caused by inefficient satellite resource allocation. Second, to address the issue of difficulty in planning for changing new task requirements in uncertain environments, an HLRS for uncertain environments is developed to ensure efficient task insertion while avoiding conflict costs. Finally, an MSFADE algorithm is presented to handle the problem of long MSIMPUE mission response time and low mission completion rate with good quality in an acceptable computation time. The simulation results validated the effectiveness and stability of the method in dealing with MSIMPUE. Moreover, the HLRS-MSFADE algorithm outperforms previous methods in terms of mission response time, mission completion rate, and motion perturbation.

To address the challenges of slow convergence and suboptimal scheduling quality in dynamic Earth observation satellite mission planning with stochastic task arrivals, this paper proposes a fusion method (D-DRL-GA) that employs an improved Genetic Algorithm (GA) to guide exploration in a Double Deep Reinforcement Learning (DRL) framework. Specifically, the GA’s generational evolution is modeled as a Markov Decision Process, and a six-dimensional state vector—is designed to fully characterize population and task-load dynamics. A Double Deep Q-Network (DDQN) agent then dynamically adjusts crossover rate, mutation rate, and operator selection to adaptively optimize GA parameters and operators online, while the diverse task sequences generated by the GA guide the DRL agent’s exploration, significantly enhancing global search capability and convergence efficiency in high-dimensional action spaces. To simulate emergency task arrivals, two injection mechanisms—burst and sporadic—are introduced, improving the system’s responsiveness to sudden tasks. In simulation experiments, D-DRL-GA outperforms Standard GA, Adaptive GA, Memetic Algorithm (MA), and NSGA-II, achieving approximately 3.5% higher total reward, 2.8% higher completion rate, and over 10% reduction in scheduling time, thereby validating its high efficiency and robustness in multi-constraint dynamic mission planning.

Mission scheduling is an essential function of the management control of remote-sensing satellite application systems. With the continuous development of remote-sensing satellite applications, mission scheduling faces significant challenges. Existing work has many inherent shortcomings in dealing with dynamic task scheduling for remote-sensing satellites. In high-load and complex remote sensing task scenarios, there is low scheduling efficiency and a waste of resources. The paper proposes a scheduling method for remote-sensing satellite applications based on dynamic task prioritization. This paper combines the and Bound methodologies with an onboard task queue scheduling band in an active task prioritization context. A purpose-built emotional task priority-based scheduling blueprint is implemented to mitigate the flux and unpredictability characteristics inherent in the traditional satellite scheduling paradigm, improve scheduling efficiency, and fine-tune satellite resource allocation. Therefore, the Branch and Bound method in remote-sensing satellite task scheduling will significantly save space and improve efficiency. The experimental results show that comparing the technique to the three heuristic algorithms (GA, PSO, DE), the BnB method usually performs better in terms of the maximum value of the objective function, always finds a better solution, and reduces about 80% in terms of running time.

No abstract available

With the gradual increase in the breadth and depth of remote sensing satellite applications, the continuous observation of targets by multi-satellite coordinated relay has become an important means to improve information assurance capabilities. application. This paper proposes a multi-level guided satellite mission collaborative planning algorithm, and designs a modular process and a final plan verification algorithm for multi-level guidance. At the same time, this paper proposes and standardizes various constraints and inspection procedures in guided planning, and ensures the accuracy and robustness of the planning scheme by introducing effectiveness and delay constraints, uniqueness constraints and energy balance constraints. It can achieve continuous observation of various targets under the premise of meeting the timeliness requirements. The validity and accuracy of the algorithm are verified by the simulation calculation of reconnaissance satellites and point targets, regional targets, and moving targets, and the calculation planning for complex task requirements is realized.

This study was conducted in response to the challenges posed by the heterogeneity of ground station resources and the dynamic nature of tasks in satellite observation missions. To combat these issues, we propose a resource scheduling method based on a dynamic coalition algorithm. The method involves constructing a five-dimensional evaluation system including spatial proximity, energy sufficiency, equipment integrity, load balancing, and continuous observation capability, which is combined with an improved simulated annealing algorithm to achieve global optimization of the coalition structure. Then, an energy allocation strategy based on demand is designed to enhance system sustainability. An experiment comparing the greedy, particle swarm, genetic, and simulated annealing algorithms was conducted. The results showed that the task completion rate of the dynamic coalition algorithm reached 93.8%; the resource utilization rate was 85.7%; the energy consumption standard deviation was 18.7; and the convergence speed was 45 iterations for the proposed method. These results were significantly better than those of other algorithms used for comparison. The innovative aspects of this study include ① a dynamic energy allocation model based on normalized priority; ② a simulated annealing optimization framework with hybrid neighborhood operations; and ③ the deep integration of multi-dimensional evaluation metrics and dynamic coalition construction mechanisms. This research provides theoretical support and technical solutions for task scheduling in wireless sensor networks under complex dynamic scenarios.

Satellite telemetry, tracking, and command (TT&C) technology plays a critical role in maintaining stable operation and emergency scheduling in mega-satellite networks. However, due to the high-speed movement of satellites, the intermittent connection between the satellite and the ground station creates a dynamic and complex visibility period, leading to temporal resource utilization conflicts during the TT&C mission scheduling process. These conflicts can further exacerbate real-time response for emergency TT&C missions in large-scale satellite networks. To address this issue, we explore the time-aware dynamic characteristics of emergency TT&C missions and their impact on the scheduling time of routine missions. Then we propose an efficient method for reducing resource utilization conflicts based on Dynamic Priority and Minimum Disturbance (DPMD). Building on this method, we formulate the scheduling process of emergency TT&C missions as a reinforcement learning decision problem suitable for dynamic real-time environments. Additionally, we introduce the DRLETMS algorithm (Emergency TT&C Missions Scheduling Algorithm based on Deep Reinforcement Learning) to achieve faster mission scheduling strategies in highly dynamic environments. Simulations demonstrate the superior efficiency of our proposed algorithm in large-scale scenarios compared to typical heuristic algorithms. Furthermore, we examine the impact of various ground station distributions on the real-time TT&C performance of satellite networks under the same satellite scale, which can provide theoretical guidance for designing mega-satellite TT&C networks in the future.

Breaking the barriers of the ubiquitous “one type of satellite, one system” resource management is an important exploration to enhance the effectiveness of collaborative work of various satellites in satellite networks. This means that flexible cross-domain mission scheduling (CMS) between different service domains becomes possible. However, the coexistence of various missions with differentiated requirements is challenging for efficient collaboration of resources in different service domains to ensure CMS performance. To this end, this paper first establishes a general mission requirement representation model to systematically characterize the differences of various missions to provide an intuitive decision-making reference for the CMS. Hereafter, based on the model, the CMS problem of coupling inter-domain mission transfer and intra-domain resource competition is solved by developing a multi-agent reinforcement learning-based hierarchical intelligent CMS algorithm. The proposed algorithm can dynamically adjust and match the CMS policy according to different mission requirements to obtain performance gain of cross-domain transmission of missions. Besides, the design of an invalid action filter and a collaboration mechanism provides efficient guidance for the CMS under resource and mission requirement constraints. Simulation results demonstrate the significant performance improvement of the proposed algorithm compared with independent domains and the existing CMS algorithms, and it can still guarantee high service performance under different network scales.

Satellite emergency mission scheduling scheme group decision making (SEMSSGDM) is a key part of satellite mission scheduling research. An appropriate evaluation model can provide a dependable and sustainable improvement and guide the functioning of emergency mission scheduling. Consequently, this research is devoted to proposing a novel decision-making method that employs a novel consensus model with hesitant fuzzy 2-tuple linguistic sets (HF2TLSs) to eliminate disagreements among satellite dispatchers and reach consensus in scheme decision-making. Within the novel method, it proposes a distance measurement function based on Hausdorff distance with HF2TLS to gauge the fit and similarity across satellite dispatchers. Additionally, a consensus reaching process (CRP) is designed to adjust the judgement of satellite dispatchers taking into account the trust degree to improve consensus. Within the selection process, a combination of the particle swarm optimization (PSO) algorithm and the MULTIplicative MOORA (MULTIMOORA) method is applied, where PSO is performed to improve the accuracy of information aggregation, and the MULTIMOORA method is used to develop the robustness of the selection results. Lastly, an applicative example validates the effectiveness of the method based on a mission scheduling intelligent decision simulation system.

No abstract available

Efficient and adaptive mission planning for Earth Observation Satellites (EOSs) remains a challenging task due to the growing complexity of user demands, task constraints, and limited satellite resources. Traditional heuristic and metaheuristic approaches often struggle with scalability and adaptability in dynamic environments. To overcome these limitations, we introduce AEM-D3QN, a novel intelligent task scheduling framework that integrates Graph Neural Networks (GNNs) with an Adaptive Exploration Mechanism-enabled Double Dueling Deep Q-Network (D3QN). This framework constructs a Directed Acyclic Graph (DAG) atlas to represent task dependencies and constraints, leveraging GNNs to extract spatial–temporal task features. These features are then encoded into a reinforcement learning model that dynamically optimizes scheduling policies under multiple resource constraints. The adaptive exploration mechanism improves learning efficiency by balancing exploration and exploitation based on task urgency and satellite status. Extensive experiments conducted under both periodic and emergency planning scenarios demonstrate that AEM-D3QN outperforms state-of-the-art algorithms in scheduling efficiency, response time, and task completion rate. The proposed framework offers a scalable and robust solution for real-time satellite mission planning in complex and dynamic operational environments.

The number of real-time dynamic satellite observation missions has been rapidly increasing recently, while little attention has been paid to the dynamic mission-scheduling problem. It is crucial to reduce perturbations to the initial scheduling plan for the dynamic mission-scheduling as the perturbations have a significant impact on the stability of the Earth observation satellites (EOSs). In this paper, we focus on the EOS dynamic mission-scheduling problem, where the observation profit and perturbation are considered simultaneously. A multi-objective dynamic mission-scheduling mathematical model is first formulated. Then, we propose a multi-objective dynamic mission-scheduling algorithm (MODMSA) based on the improved Strength Pareto Evolutionary Algorithm (SPEA2). In the MODMSA, a novel two-stage individual representation, a minimum perturbation random initialization, multi-point crossover, and greedy mutation are designed to expand the search scope and improve the search efficiency. In addition, a profit-oriented local search algorithm is introduced into the SPEA2 to improve the convergence speed. Furthermore, an adaptive perturbation control strategy is adopted to improve the diversity of non−dominated solutions. Extensive experiments are conducted to evaluate the performance of the MODMSA. The simulation results show that the MODMSA outperforms other comparison algorithms in terms of solution quality and diversity, which demonstrates that the MODMSA is promising for practical EOS systems.

No abstract available

Satellite formation flying with functional payloads can provide flexibility and real-time services for time sensitive missions through the cooperation of intra-satellite wired and inter-satellite wireless communication systems. However, the difference between wired and wireless scheduling mechanisms will increase the instability of forwarding delay at satellite. To solve this problem, we propose a wired and wireless converged scheduling scheme. Firstly, we construct the transmission rate and time division multiple access scheduling model to describe the inter-satellite link, and the IEEE 802.1Qbv scheduling model for wired link respectively. Secondly, we analyze the influence of the position of packet in different time slots on the forwarding delay, to model the relationship between the converged scheduling and forwarding delay. Finally, we formulate the converged scheduling as a zero-one programming problem with minimum jitter, and adopt the genetic-tabu search algorithm to solve it. Simulation results demonstrate that the delay is reduced by 20%, the jitter is not higher than $40\mu\mathrm{s}$.

Data Relay Satellite Networks (DRSNs) face the challenge of balancing increasing relay mission demands with limited resources. Dynamic occurrences further complicate this challenge, particularly in addressing large-scale problems. Therefore, this paper proposes an end-to-end deep reinforcement learning (DRL) approach to maximize total task profits within dynamic DRSN environments. A continuous-time Markov decision process model is formulated to effectively represent the problem, significantly reducing the length of the solution sequence without degrading scheduling quality. Additionally, an interactive training framework comprising two phases—pre-training and intensive training—is proposed to enhance the model's performance and generalization capability. In both phases, populations are employed instead of single-shot inference, providing a more efficient approach to dynamic environment exploration. Once trained, the policy network can solve problems of varying sizes in static and dynamic benchmark tests. Experimental results demonstrate that the proposed method, with its strong generalization capability across different problem sizes, outperforms state-of-the-art DRL and heuristic methods regarding total task profits, running time and completion rate.

Agile earth observation satellites (AEOS) play a critical role in improving our understanding and real-time sensing of the Earth and its environment. In recent years, as the demand for satellite observation and the number of missions have increased dramatically, the agile earth observation satellite mission scheduling problem (AEOSSP) has become increasingly demanding in terms of computational efficiency and algorithmic performance for large-scale solutions. In this paper, we propose a scheduling method based on reinforcement learning to address the above challenges. First, we model the problem as a multidimensional multi-knapsack problem with conflicts. Then, we build a neural network and design the reward function to reflect the problem-specific objectives and the penalty function to reflect the constraints. We train this network using the actorcritic model within reinforcement learning, combining it with the policy gradient method to directly optimize the total task gain. Finally, the trained model outputs a probability vector, which the model generates, and thus the optimal task allocation results are derived. In order to verify the effectiveness of the proposed algorithm, multiple sets of test cases are generated by combining the Satellite Tool Kit (STK) simulation. The related test results show that the proposed algorithm is better than the classical genetic algorithm in terms of algorithmic efficiency, task gain and its stability index, which verifies its computational efficiency in solving this data set.

The optimization of satellite planning to improve system performance has always been a topic of interest. The problem of planning a remote-sensing satellite involves selecting and scheduling tasks from a set of user requests to optimize one or more objective functions. In recent years, artificial intelligence has proven to be an effective planning tool for remote-sensing satellites. This paper investigates the planning of an observation mission by a remote-sensing satellite, with the goal of optimizing two objective functions (failure rate and timely execution of missions) using artificial intelligence. Reinforcement learning and multi-objective optimization using the Pareto front are used as the solving method. This research considers the dynamic characteristics of energy, memory, and attitude in addition to its baseline research characteristics. By using this method, several optimal agents have been identified. The results indicate that integrating realistic operational constraints, such as memory usage, updated databases, and power/transition management, into the optimization framework reduces the solution space but leads to more practical and deployable outcomes. A clear tradeoff exists between solution diversity and realism. While incorporating operational factors restricts optimization flexibility, it simultaneously improves the applicability and reliability of the solutions.

The efficient scheduling of Earth observation tasks for remote sensing satellites, each with diverse resolution and delivery time constraints, remains a formidable challenge. This paper proposes a novel digital twin model integrated with a genetic algorithm (GA) to address the complex multi-request and multi-constraint scheduling problem. The proposed GA-based digital twin model incorporates satellite orbital data, ground station resources, and communication/imaging windows. The proposed model automatically generates near-optimal mission planning schedules that minimize tardiness while satisfying stringent imaging requirements. Moreover, the model features dynamic multi-hop uplink/downlink routing and adaptive resource utilization, achieving scheduling optimizations within 150−220 seconds per request. It also enhances scalability and flexibility in real-world operations. This paper underscores the potential of digital twin technology in revolutionizing satellite mission planning and operational scheduling by significantly reducing computational overhead and the need for human intervention.

No abstract available

No abstract available

The rapid growth in satellite internet services, exemplified by constellations like Starlink, has underscored the imperative for efficient service allocation in satellite networks. This poses significant challenges due to the dynamic and multidimensional nature of user demands, which vary across time and space, alongside the fluctuating availability of satellite resources. Traditional static allocation methods are inadequate for these dynamic conditions, often resulting in inefficient resource utilization and compromised service quality. To address these challenges, this paper presents Demand-Oriented Robust Service Allocation (DORSA) with a multi-phase task matching approach. Our method begins with a static matching phase to establish an initial user demand-service match, followed by a dynamic matching phase that leverages a sparse time-varying bipartite graph model to enhance computational efficiency. The DORSA algorithm incorporates local backup-aware matching to rapidly adapt to demand fluctuations and satellite movements, ensuring continuous and high-quality service provision. Simulation results indicate that our approach reduces the average response latency by 74.5% compared to the baseline approach, and by 54.2% compared to the state-of-the-art task allocation algorithm under dynamic demand conditions within satellite networks.

This paper proposes a dynamic link allocation strategy for multi-layer satellite networks (MLSNs) to address link congestion and load imbalance in the Low Earth Orbit (LEO) layer. By introducing dynamic and stable inter-orbital links (D-IOLs and S-IOLs), the algorithm dynamically adjusts link allocation to respond to sudden traffic changes while ensuring network stability. A greedy algorithm is employed for dynamic central satellite selection, while stable central satellites are optimized based on visibility time. The proposed framework achieves a balance between flexibility and stability by leveraging the complementary advantages of D-IOLs and S-IOLs. Additionally, global load balancing further enhances system performance by ensuring equitable traffic distribution across the network. Simulation results demonstrate that the proposed algorithm significantly reduces link congestion and optimizes load distribution compared to traditional schemes and service-time-based schemes. Notably, the proposed method exhibits superior adaptability, load balancing, and congestion mitigation under high-load scenarios, offering valuable insights for advancing satellite communication systems.

This paper investigates dynamic task allocation for multi-agent systems (MASs) under resource constraints, with a focus on maximizing the global utility of agents while ensuring a conflict-free allocation of targets. We present a more adaptable submodular maximization framework for the MAS task allocation under resource constraints. Our proposed distributed greedy bundles algorithm (DGBA) is specifically designed to address communication limitations in MASs and provides rigorous approximation guarantees for submodular maximization under $q$-independent systems, with low computational complexity. Specifically, DGBA can generate a feasible task allocation policy within polynomial time complexity, significantly reducing space complexity compared to existing methods. To demonstrate practical viability of our approach, we apply DGBA to the scenario of active observation information acquisition within a micro-satellite constellation, transforming the NP-hard task allocation problem into a tractable submodular maximization problem under a $q$-independent system constraint. Our method not only provides a specific performance bound but also surpasses benchmark algorithms in metrics such as utility, cost, communication time, and running time.

No abstract available

Currently, most resource allocation algorithms in satellite networks only focus on single resource allocation scenarios. But there are many kinds of resources in the actual satellite network, and these resources restrict each other. In addition, most satellite network systems treat the services requested by users as an independent and indivisible task. However, the duration of the task is different and satellites have time windows to execute tasks, as well as multiple resource constraints. With the increase of tasks, the number of tasks that the satellite network system can complete and the resources utilization will be limited. For the above problems, this paper proposes a multi-dimensional resource allocation algorithm based on task splitting and adjustment, including task splitting algorithm (TSA) and dynamic task adjustment algorithm (DTAA). First, build a model that takes task completion rate, resource utilization and distriction window utilization as the optimization targets in a single distriction window, and solve the problem of lower resource utilization by splitting the last task that cannot be executed in the distriction window. Then, build a dynamic task adjustment model among multiple distriction windows, and solve the conflict problem of task execution through the original window task adjustment or the new window task adjustment strategy. Finally, compare the algorithm with the existing decoupled resource algorithm (DRA) and improved greedy algorithm (IGA), and verify its gains in the number of tasks completed, resource utilization and distriction window utilization through simulation experiments.

Multi-access edge computing (MEC) will play an important role in future Internet of Thing (IoT) networks. However, MEC-enabled systems depend heavily on telecommunication infrastructure and network access, which often unavailable in emergency scenarios and hot spots. In view of this, the delivery of service data from massive IoT sensor devices is so important that it should be solved. In this paper, we propose a MEC-empowered satellite-terrestrial network (STN) assisted through millimeter wave (mmWave) small-cell base stations (SBSs) and low earth orbit (LEO) satellite, in which the data collected by sensor devices can be delivered to STN nodes for processing directly. Our main objective is to minimize the total task latency for two types of sensing services with diverse quality-of-service (QoS) demands. Since the formulated optimization problem is a mixed-integer non-linear programming (MINLP) problem, which is NP-hard, we further divide the joint problem into three subproblems and adopt swap matching-based alternative iteration method to reduce the computation complexity. Numerical results prove that our proposed method outperforms the benchmarks and can achieve up to 99.57% performance of the exhaustive search method.

Active space debris removal is now integral to modern space exploration. In order to address the problem of a heterogeneous satellite swarm with different payloads carrying out the emergency active removal of space debris, this paper proposes a Multi-type Chromosome Fast Elitist Non-Dominated Sorting Genetic Algorithm (MC-NSGA-II). The algorithm is designed to enable the satellite swarm to execute multiple coupled tasks in succession with improved optimization efficiency. An arbitrary execution order may result in deadlock, where one or more satellites become trapped in an infinite waiting loop. In order to address the heterogeneous problem of satellites and task coupling constraints, a multi-type chromosome coding strategy is developed. To evaluate different allocation strategies, three optimization objectives—time consumption, fuel consumption, and task balance—are introduced. To align with the multi-type chromosome coding strategy, two distinct sorting methods are developed for crossover and mutation operations, ensuring that all offspring individuals meet the constraints. Additionally, the algorithm incorporates a dynamic parameter-setting strategy to enhance solution efficiency. Finally, comparative simulations validate the effectiveness and superiority of the proposed method. The results show that the high-quality solution search ability of the MC-NSGA-II algorithm is 23.07% higher than that of the standard NSGA-II algorithm.

To address the challenges of manual dispatch reliance, poor dynamic adaptability, and low collaborative efficiency in drone-assisted search and rescue operations during sudden regional disasters, this paper proposes a two-stage "bid and dispatch" allocation algorithm. The approach aims to minimize both the number of drones required and the total task completion time. The algorithm transforms search and rescue missions into standardized orders. In the bid phase, drones autonomously evaluate tasks based on their current status and task characteristics, calculating corresponding competitiveness scores. During the dispatch phase, the algorithm performs global optimization to select the most suitable drones from the candidate pool. Incorporated with multi-dimensional evaluation and dynamic reallocation mechanisms, the method ensures operational efficiency. Experimental results demonstrate that, compared to centralized allocation methods, the proposed algorithm improves response speed by over 30% and increases resource utilization by 25%, thereby effectively enhancing rescue efficiency and optimizing resource allocation.

No abstract available

No abstract available

Resource scheduling technology as an important means to achieve the optimal allocation of SAGIN resources, plays a vital role in Earth observation, emergency communications and other fields, through the in-depth understanding and analysis of the research status of the integrated network resource scheduling field, the current resource scheduling technology is not yet perfect, it is difficult to meet the multi-satellite collaboration, network topology real-time changes, massive users as the characteristics of the scheduling needs. Starting from the research point of multi-dimensional resources, this paper considers the synergy between resources, proposes a multi-dimensional resource vector model, and on this basis proposes a multi-dimensional resource collaborative optimization scheduling algorithm to achieve unified scheduling from the global perspective. In the simulation experiment, it is proved that the indicators such as the number of task completions, total tasks revenues, average revenue and scheduling decision-making time can be effectively improved, among which the number of task completions is increased by 10%-30%, the total tasks revenue is increased by 30%-60%, and the scheduling decision-making time is saved by about 30%.

With the rapid development of large-scale low-Earth orbit (LEO) satellite Internet, very LEO (VLEO) Earth observation constellations are increasingly using intersatellite links (LISLs) for cross-layer access to Internet satellites. It is an effective means to enhance data return throughput. When both observation and communication satellite constellations use lasers for networking, cross-layer access between the VLEO and LEO satellite networks requires reallocating lasers. This reallocation disrupts the original topology and impacts network performance. To maximize the collaborative operational efficiency of the double-layer network, we fundamentally analyze the relationship between topology links, throughput, and delay using graph and queuing theory. Innovatively, we discover the superiority of two axioms in addressing the cross-layer topology optimization problem, including the “Minimum Hop Count” and “Minimum Overlap Path.” Based on these axioms, a many-objective cross-layer topology optimization model is established that considers the hop count and the average link utilization frequency of both VLEO and LEO satellite networks. To reduce the reliance on centralized algorithms on global transmission demand, a local distributed interaction mechanism (LDIM) is proposed for cross-layer LISL establishment. An onboard novel distributed many-objective cross-layer topology optimization (NDMTO) algorithm is also introduced for VLEO satellites to manage access strategies. Finally, we use real data from Typhoon LEKIMA to create a multitask scenario and conduct packet-level simulations based on the Starlink and Dove constellations. The results indicate that, compared to existing benchmarks, the NDMTO algorithm improves data throughput by 26.73% and reduces the average transmission delay of emergency task data by 20.7%.

No abstract available

In recent years, the demand for remote sensing satellite applications has grown rapidly, with diversified observation needs from users at all levels presenting an unprecedented scale and complexity. Traditional satellite task planning methods have obvious limitations in task allocation, scheduling, and resource assignment, making it difficult to achieve efficient collaboration and precise decision-making, and failing to effectively adapt to the growing observation demands and increasing complexity. To address these challenges, there is an urgent need to develop intelligent star-ground coordinated observation task planning methods. This article proposes a satellite-ground coordinated satellite observation task intelligent planning method based on reinforcement learning. It uses neural networks to intelligently predict observation object information to generate satellite observation demands and employs neural networks through intelligent agents for smart decision-making of satellite observation tasks, automatically generating satellite observation task timelines. The satellite task planning system and the ground task planning system use the same deep learning network model. Historical data is used on the ground to train the neural network, and the trained model parameters are then uploaded to the satellite, achieving intelligent planning of satellite-ground coordinated satellite observations. An LSTM network model is adopted for predicting the longitude and latitude coordinates of observation objects. Experimental results show that the model exhibits rapidly decreasing loss values during training with good convergence performance. A framework combining reinforcement learning algorithms with MDP (Markov Decision Process) is constructed to address the smart scheduling problem of multiple observation objects. Experiments demonstrate that this method can find optimal decision paths under multiple constraints, aligning with expectations. Research indicates that this intelligent planning method integrating LSTM and MDP holds promising applications in satellite observation task planning, providing efficient solutions for practical applications.

Agile satellites leverage rapid and flexible maneuvering to image more targets per orbital cycle, which is essential for time-sensitive emergency operations, particularly disaster assessment. Correspondingly, the increasing observation data volumes necessitate the use of on-orbit computing to bypass storage and transmission limitations. However, coordinating precedence-dependent observation, computation, and downlink operations within limited time windows presents key challenges for agile satellite service optimization. Therefore, this paper proposes a deep reinforcement learning (DRL) approach to solve the joint observation and on-orbit computation scheduling (JOOCS) problem for agile satellite constellations. First, the infrastructure under study consists of observation satellites, a GEO satellite (dedicated to computing), ground stations, and communication links interconnecting them. Next, the JOOCS problem is described using mathematical formulations, and then a partially observable Markov decision process model is established with the objective of maximizing task completion profits. Finally, we design a joint scheduling decision algorithm based on multiagent proximal policy optimization (JS-MAPPO). Concerning the policy network of agents, a problem-specific encoder–decoder architecture is developed to improve the learning efficiency of JS-MAPPO. Simulation results show that JS-MAPPO surpasses the genetic algorithm and state-of-the-art DRL methods across various problem scales while incurring lower computational costs. Compared to random scheduling, JOOCS achieves up to 82.67% higher average task profit, demonstrating enhanced operational performance in agile satellite constellations.

A multi-target robust observation method for satellite constellations based on hypergraph algebraic connectivity and observation precision theory is proposed to address the challenges posed by the surge in space targets and system failures. First, a precision metric framework is constructed based on nonlinear batch least squares estimation theory, deriving the theoretical precision covariance through cumulative observation matrices to provide a theoretical foundation for tracking accuracy evaluation. Second, multi-satellite collaborative observation is modeled as an edge-dependent vertex-weighted hypergraph, enhancing system robustness by maximizing algebraic connectivity. A constrained simulated annealing (CSA) algorithm is designed, employing a precision-guided perturbation strategy to efficiently solve the optimization problem. Simulation experiments are conducted using 24 Walker constellation satellites tracking 50 targets, comparing the proposed method with greedy algorithm, CBBA, and CSA-bipartite Graph methods across three scenarios: baseline, maneuvering, and failure. Results demonstrate that the CSA-hypergraph method achieves 0.089 km steady-state precision in the baseline scenario, representing a 41.4% improvement over traditional methods; in maneuvering scenarios, detection delay is reduced by 34.3% and re-achievement time is decreased by 47.4%; with a 30% satellite failure rate, performance degradation is only 9.8%, significantly outperforming other methods.

Agile satellite imaging scheduling plays a vital role in improving emergency response, urban planning, national defense, and resource management. With the rise in the number of in-orbit satellites and observation windows, the need for diverse agile Earth observation satellite (AEOS) scheduling has surged. However, current research seldom addresses multiple optimization objectives, which are crucial in many engineering practices. This article tackles a multiobjective AEOS scheduling problem (MOAEOSSP) that aims to optimize total observation task profit, satellite energy consumption, and load balancing. To address this intricate problem, we propose a strategy-fused multiobjective dung beetle optimization (SFMODBO) algorithm. This novel algorithm harnesses the position update characteristics of various dung beetle populations and integrates multiple high-adaptability strategies. Consequently, it strikes a better balance between global search capability and local exploitation accuracy, making it more effective at exploring the solution space and avoiding local optima. The SFMODBO algorithm enhances global search capabilities through diverse strategies, ensuring thorough coverage of the search space. Simultaneously, it significantly improves local optimization precision by fine-tuning solutions in promising regions. This dual approach enables more robust and efficient problem-solving. Simulation experiments confirm the effectiveness and efficiency of the SFMODBO algorithm. Results indicate that it significantly outperforms competitors across multiple metrics, achieving superior scheduling schemes. In addition to these enhanced metrics, the proposed algorithm also exhibits advantages in computation time and resource utilization. This not only demonstrates the algorithm’s robustness but also underscores its efficiency and speed in solving the MOAEOSSP.

Periodic observation is widely used for information collection in emergency scenarios such as earthquakes, wildfires, and floods. Multi-satellite observation is needed to achieve pe-riodic observation due to the limited observation capability and long revisit period of a single satellite. Choosing when and which satellite to observe the target to achieve periodical observation can be challenging. Existing efforts adopt a time-consuming method to solve the problem. They also ignore some feasible solutions. Moreover, the classical mathematical programming method can also take a long time to solve the problem. Therefore, we propose a graph-based periodic scheduling method. First, we formulate the problem as a Mixed-Integer Linear Programming (MILP) problem, which is non-convex and intractable. To address this, we introduce a periodic observation graph model that transforms the original problem into a path-finding problem. We then propose a graph-based periodic observation path construction algorithm. Numerical simulations under the settings of a realworld satellite network demonstrate that the performance of our scheme is significantly faster than the baselines and can obtain observation offsets.

With the rapid development of large‐scale satellite constellations and the intelligent progression of satellites, distributed task planning algorithms have garnered extensive research. However, traditional distributed algorithms face drawbacks such as increased communication burdens and slower response to emergency tasks when dealing with task planning for large satellite clusters. This paper introduces improvements to the traditional contract network protocol to address these shortcomings. Firstly, a method for satellites to determine their own dominant tasks is proposed with the aim of alleviating the superfluous negotiation communication among satellites. Secondly, in the context of batch task upload scenarios, a multitask single‐bid methodology is designed. Furthermore, a provisional task group was constituted to conduct small‐scale intersatellite negotiations for exceptional and urgent tasks for the purpose of minimizing the response time of the system. The simulation results indicate that the algorithm marginally enhances the computing memory of satellites but decreases the intersatellite communication traffic.

As a driving force in the advancement of intel-ligent in-orbit applications, DNN models have been gradually integrated into satellites, producing daily latency-constraint and computation-intensive tasks. However, the substantial computation capability of DNN models, coupled with the instability of the satellite-ground link, pose significant challenges, hindering the timely completion of tasks. It becomes necessary to adapt to task stream changes when dealing with tasks requiring latency guarantees, such as dynamic observation tasks on the satellites. To this end, we consider a system model for a collaborative inference system with latency constraints, leveraging the multi-exit and model partition technology. To address this, we propose an algorithm, which is tailored to effectively address the trade-off between task completion and maintaining satisfactory task accuracy by dynamically choosing early-exit and partition points. Simulation evaluations show that our proposed algorithm signif-icantly outperforms baseline algorithms across the task stream with strict latency constraints.