40 results about "Satellite scheduling" patented technology

The invention relates to a simulation-based satellite scheduling task dynamic planning method, device and equipment, and the method comprises the steps: carrying out the modeling according to various time-varying environment dynamic parameters in an environment where a satellite is located, and obtaining a dynamic environment model which comprises an environment factor model and an emergency triggering mechanism; the method comprises the following steps: constructing a satellite resource dynamic model according to various resources of satellites, resource consumption during task execution and resource supply, constructing a multi-satellite collaborative model according to an interaction relationship among a plurality of satellites and related information of each satellite, and constructing a multi-satellite collaborative model under a dynamic environment model, the satellite resource dynamic model and the multi-satellite collaborative model. According to the method, tasks in a to-be-executed task data set containing single-satellite and cooperative tasks are dynamically scheduled by adopting a task planning algorithm, a task execution planning table of satellites is obtained, satellite task planning can be carried out aiming at emergencies and multi-satellite cooperative tasks, and the planning is closer to the actual situation.

The embodiment of the invention provides a satellite beam scheduling method and device based on reinforcement learning, equipment and a storage medium. The method comprises the following steps: constructing a state space comprising respective satellite identifiers of a plurality of satellites in a navigation satellite constellation, pitch angles of the satellites relative to a target area, direction angles of the satellites relative to the target area and relative distances between the satellites and the target area, and constructing an action space comprising a satellite identifier sequence of satellites to be scheduled, the method comprises the steps of designing a reward function based on a reward index comprising a coverage multiple reward and a PDOP reward, training a neural network of satellite beam scheduling according to a state space, an action space and the reward function based on a TD3 algorithm to obtain a trained strategy network, inputting a real-time state vector of a target area into the trained strategy network to obtain a real-time action, and outputting the real-time action to the target area. And controlling all satellite beams corresponding to the real-time action to point to the target area. The method can improve the service quality of the satellite and improve the accuracy of reentry trajectory prediction.

The invention discloses a multi-satellite task allocation strategy adaptive selection method based on deep reinforcement learning, and aims to solve the problems that the satellite task scheduling problem is difficult to solve due to the fact that the satellite attitude maneuver ability is improved, and the task number and the satellite scale are explosively increased, and the solving limitation of a traditional method is solved. According to the method, the problem is divided into an upper-layer multi-satellite task allocation part and a lower-layer single-satellite scheduling part, iterative optimization of a scheme is realized through multiple interactions of the two parts, a self-adaptive search method based on reinforcement learning is provided for the upper-layer task allocation problem, and task allocation strategy self-adaptive selection based on deep reinforcement learning is realized.

The invention relates to a method, system and device for scheduling tasks of low-orbit constellations for dynamic targets in space under a fully distributed satellite system, belonging to the field of on-orbit mission planning technology for spacecraft, and solving the problems of high technical cost, low flexibility and low stability of low-orbit constellation task scheduling. The method of the invention comprises: using an autonomous distributed task scheduling method based on CCGA to solve the multi-satellite scheduling problem, decomposing the multi-satellite scheduling problem into a single-satellite scheduling problem, and solving the optimal scheduling solution for each satellite separately to form an overall optimal scheduling solution. The scheduling process consists of local optimization and global negotiation. Each satellite independently explores the scheduling problem of the satellite for local optimization. At the same time, a ring topology is used to form an inter-satellite communication network for global negotiation within the constellation. The observation effect of the scheduling solution is comprehensively evaluated by the weighted sum of the satellite contribution, the number of switching times and the satellite load. The task benefit evaluation index is introduced to assist in the optimization of the single-satellite scheduling solution. The invention is suitable for low-orbit constellation task scheduling.

The invention discloses a resource scheduling optimization method for a giant constellation, relates to the technical field of satellite resource allocation of the giant constellation, solves the problem that an existing satellite resource scheduling method is difficult to give consideration to global optimization of time and energy consumption, and aims to improve the execution efficiency of a multi-satellite cooperative task and reduce energy consumption. A real satellite scheduling task is simulated and arranged to obtain satellite data and task data, an evolutionary process of a reinforcement learning guide genetic algorithm is adopted to optimize a task sequence, satellite resource allocation is performed on the optimized task sequence, time and energy consumption of the task sequence are completed, and finally normalization is performed to obtain a comprehensive index. According to the method, a task execution sequence with the minimum time and energy consumption comprehensive index is obtained, so that an optimal scheduling scheme is obtained, and the requirement of satellite scheduling under a giant constellation is met.

The invention provides an inter-satellite communication service scheduling method and system for low-orbit satellites, a medium and a terminal. An edge server is deployed on each satellite; the method comprises the following steps: calculating a satellite comprehensive weight of each satellite in a time slot; calculating a cache path collaborative weight in the time slot, so as to pre-cache high-frequency access data of inter-satellite communication on the edge server based on the cache path collaborative weight; and realizing inter-satellite scheduling of multiple tasks based on a directed acyclic graph driving algorithm and the high-frequency access data. According to the inter-satellite communication service scheduling method and system for the low-earth-orbit satellites, the medium and the terminal, the bandwidth utilization efficiency of inter-satellite communication of the low-earth-orbit satellites is effectively improved based on the combination of the edge server and DAG driving, and the time delay of communication task processing is greatly reduced.

The invention provides a design method and system for a task planning algorithm digital parallel verification platform, and the design method is based on an existing digital twin system, restores a real application environment of a task planning algorithm, and achieves the full-process simulation of the application effect of the algorithm in a real environment. According to the method, a universal interface protocol for algorithm integration is provided, so that any algorithm conforming to the protocol can be called by a platform. For a plurality of task planning algorithms, the method provides a Restful-based hot plug strategy so as to realize random switching of the plurality of algorithms in the same scene. After the task planning algorithm is called, the method further provides an evaluation scheme for the satellite scheduling result, and the evaluation scheme is used for evaluating the execution effect of the algorithm. According to the method, powerful support is provided for development, test and optimization of a low earth orbit satellite task planning algorithm, a solid foundation is laid for solving many complex problems in low earth orbit satellite task scheduling, and the requirement for efficient and stable operation of a low earth orbit satellite system in a complex environment is met.

The invention relates to the technical field of satellite scheduling, in particular to a satellite-borne resource scheduling method in a multi-task coordination scene, which comprises the following steps: S1, constructing a single-satellite multi-task coordination scene; s2, judging the priority of a to-be-completed task of a satellite based on a multi-task cooperation scene, and analyzing evaluation indexes of different tasks and satellite-borne resources required by the evaluation indexes; and S3, designing a spaceborne resource scheduling algorithm based on spaceborne resources required by different tasks. According to the satellite-borne resource scheduling method in the multi-task collaborative scene, an evolution model based on a dynamic environment and a grey correlation analysis method are adopted, the diversity of satellite-borne resource types in the modeling process can be kept, meanwhile, the importance degrees of the tasks are ranked, and the task priorities are determined; and designing a corresponding spaceborne resource scheduling algorithm, and finally constructing a spaceborne resource scheduling method in a multi-task collaborative scene, so that reasonable allocation, efficient management and dynamic adjustment of spaceborne resources can be realized.

The invention discloses a relay satellite military and civilian fusion multi-target scheduling method based on deep reinforcement learning. The method specifically comprises the following steps: step 1, collecting data of a user task set and a relay satellite resource set; step 2, establishing a mathematical model according to the data collected in the step 1 by following a relay satellite scheduling principle; step 3, realizing a process of scheduling the mathematical model established in the step 2 through a Markov decision process; and step 4, optimizing the decision process in the step 3 through deep reinforcement learning to realize relay satellite military-civilian fusion multi-target scheduling. The problems that the scheduling center conflicts in civil and military major project scheduling, user application and execution are not timely, and the utilization rate of relay satellite resources is low are solved.

The invention relates to the technical field of space network communication and satellite intelligent management, in particular to an LEO satellite network intelligent routing system and method based on intention recognition. Comprising a natural language instruction input module, a text preprocessing module, a BERT semantic modeling module for RoPE position coding enhancement, a bidirectional LSTM context enhancement module, an intention recognition module, an entity recognition module, a structured result generation module and an interface interaction module. According to the method, a path request instruction in a natural language form is mapped into a structured scheduling parameter, the problems that a parameter configuration process in a traditional satellite scheduling system depends on manpower, the response timeliness is poor and the analysis capability is weak are solved, meanwhile, rotation position coding is introduced to improve the sequence structure sensing capability, and local context features are captured in combination with BiLSTM, so that the real-time performance of the satellite scheduling system is improved. Through intention recognition and entity recognition dual-task joint modeling, high-precision output of a structured request analysis result is realized, and key support is provided for satellite network intelligent scheduling.

The invention discloses a relay satellite scheduling method considering antenna resource heterogeneity and task diversity, and the method specifically comprises the following steps: 1, discretizing a scheduling period into equal-length time slots, and constructing a time slot set S; constructing a visibility mask and a service window mask for any task t and antenna a; 2, generating an initial feasible scheduling scheme by adopting a conflict perception initialization strategy; and step 3, performing iterative improvement on the initial scheme generated in the CAI stage by adopting a constraint local re-optimization strategy, thereby improving the task completion rate and the target function value. According to the invention, the overall task completion rate and the bandwidth utilization efficiency of the relay system can be improved.

The present application relates to the technical field of satellite scheduling, in particular to a multi-modal multi-objective satellite task scheduling method based on a ring topology, comprising the following steps: S1: taking the minimization of task request failure rate and ground station load imbalance rate as the optimization target, constructing a multi-modal multi-objective optimization scheduling model of satellite task scheduling, and determining the problem target space; S2: determining a task available time window set according to user task request, a ground station set, a satellite-ground visible time window set and a problem constraint set, and constructing a problem decision space; S3: adopting a knowledge evolutionary algorithm iterative optimization to explore and maintain more equivalent Pareto optimal solution sets, and obtaining a high-quality scheduling scheme set, wherein the knowledge evolutionary algorithm is based on a ring topology structure decision population updating strategy. The present application can realize multi-modal multi-objective optimization of the problem, so that a group of satellite task scheduling schemes considering multiple optimization targets and differentiated from each other can be obtained after a single run.

The invention relates to the technical field of satellite scheduling, in particular to a multi-modal multi-target satellite task scheduling method based on ring topology, which comprises the following steps: S1, by taking a minimum task request failure rate and a ground station load imbalance rate as optimization targets, constructing a satellite task scheduling multi-modal multi-target optimization scheduling model, and determining a problem target space; s2, determining a task available time window set according to the user task request, the ground station set, the satellite-ground visible time window set and the problem constraint set, and constructing a problem decision space; and S3, iterative optimization is carried out by adopting a knowledge evolutionary algorithm, more equivalent Pareto optimal solution sets are explored and maintained, a high-quality scheduling scheme set is obtained, and the knowledge evolutionary algorithm decides a population updating strategy based on a ring topology structure. According to the invention, multi-modal multi-objective optimization of the problem can be realized, so that a group of mutually differentiated satellite task scheduling schemes considering a plurality of optimization objectives can be obtained after single operation.

The present invention provides a satellite mission planning method, system, and device for point cluster targets. The method comprises: first, preprocessing and merging point target missions; then, determining a mission synthesis plan by extending satellite observation strips; and finally, outputting a final satellite scheduling plan by designing a satellite mission planning model to find the optimal solution. Based on the characteristics of dense point cluster target missions, the present invention implements integrated modeling and calculation by incorporating strip extension ratios into decision variables, achieving global optimization. Furthermore, the method fully considers the timeliness and profitability of the missions, enabling rapid and accurate planning of satellite point cluster target missions.

Embodiments of the present disclosure provide a satellite beam scheduling method and device based on reinforcement learning, equipment and a storage medium. The method comprises: constructing a state space comprising the satellite identifier of each satellite in a navigation satellite constellation, the pitch angle of each satellite relative to a target area, the direction angle of each satellite relative to the target area, and the relative distance of each satellite to the target area; constructing an action space comprising a satellite identifier sequence of a satellite to be scheduled; designing a reward function based on a reward index comprising a coverage multiplicity reward and a PDOP reward; training a neural network for satellite beam scheduling based on a TD3 algorithm, according to the state space, the action space and the reward function, to obtain a trained policy network; inputting a real-time state vector of the target area into the trained policy network to obtain a real-time action, so as to control all satellite beams corresponding to the real-time action to point to the target area. The method can improve the service quality of the satellite.

The invention provides a multi-satellite task scheduling method which can be applied to the technical field of task planning so as to improve the satellite resource utilization rate and the task execution efficiency. The method comprises the following steps: generating an initial satellite scheduling task of a target area according to the condition that a sub-area in the target area to be observed is covered by a satellite strip of a satellite set; filtering the satellite sub-tasks of which the degree values influenced by the meteorological factors are greater than a preset degree threshold value in the initial satellite scheduling task by utilizing the meteorological factors to obtain a meteorological optimization satellite scheduling task; performing optimization training on the meteorological optimization satellite scheduling task by using the multi-dimensional index to obtain an index optimization satellite scheduling task; generating a supplementary satellite scheduling task according to the intersection of the satellite strips of the satellites which are not called by the index optimization satellite scheduling task in the satellite set and the blank area; and scheduling the satellite set according to the index optimization satellite scheduling task and the supplementary satellite scheduling task. The invention further provides a multi-satellite task scheduling device and equipment.

The invention relates to the technical field of space remote sensing, in particular to a satellite scheduling strategy generation method and device and a storage medium, and the method comprises the steps: obtaining a plurality of imaging strips according to the imaging characteristics of the imaging load of a satellite related to a target area; wherein the target area comprises a movable target; establishing a POMDP model for the target area; generating an action mask according to the relationship between the imaging strip of the satellite and the target area; wherein the action masks comprise valid action masks and invalid action masks; and training the POMDP model based on the action mask, and generating the satellite scheduling strategy. According to the embodiment of the invention, the motion mask is applied to the solution process of the POMDP model during the multi-fall cooperative scheduling, so that the generation speed of the large-scale complex scheduling strategy of the satellite is remarkably improved, and the robustness of the scheduling strategy is relatively high.

The present application relates to an agile optical imaging satellite scheduling method and apparatus based on deep reinforcement learning. The method comprises: obtaining satellite status data and first task data; encoding the satellite status data and the first task data to obtain dynamic encoded data; performing task screening on the first static encoded data based on a local attention mechanism to obtain a task list; obtaining a task probability distribution based on the dynamic encoded data and the task list based on the local attention mechanism; determining a second task and determining the execution order of the second task based on the task list, the task probability distribution, and a preset constraint satisfaction model; deleting a task from the first static encoded data based on the second task and the satellite status data to obtain second static encoded data; repeating the above steps with the second task as the first task and the second static encoded data as the first static encoded data until the execution order of all tasks in the task list is determined; and determining a satellite scheduling plan based on the execution order in the task list.

Embodiments of the present disclosure relate to satellite switch with resynchronization. In an aspect, a terminal device receives, from a network device, an indication for determining time for at least one operation associated with switching from a first satellite to a second satellite. The at least one operation comprises at least one of (i) initiating a resynchronization procedure to the network device via the second satellite or (ii) performing an uplink transmission via the second satellite. The terminal device performs the at least one operation based on the received indication. The embodiments of the present disclosure can avoid scheduling request (SR) bursts at the t-Service and NW overloading due to scheduling a large number of PUSCH transmissions. Therefore, UEs may suffer less risk of long interruptions, and the NW may schedule first UL transmissions of UEs through the target satellite in a distributed manner.

This paper presents a resource scheduling optimization method for mega-constellations, focusing on satellite resource allocation in this field. It addresses the challenge of existing satellite resource scheduling methods in achieving global optimization while balancing time and energy consumption, aiming to improve the execution efficiency and reduce energy consumption of multi-satellite collaborative tasks. By simulating and processing real satellite scheduling tasks, satellite and task data are obtained. A reinforcement learning-guided genetic algorithm is used to optimize task sequences. These optimized sequences are then used for satellite resource allocation, and the time and energy consumption of each task sequence are calculated and normalized to obtain a comprehensive index. This method yields the task execution sequence with the minimum combined time and energy consumption index, thus providing the optimal scheduling scheme to meet the satellite scheduling requirements of mega-constellations.

The application discloses a kind of forest fire monitoring methods based on low-orbit satellite, comprising: obtaining data by multiple sensors, and generating fusion dataset by denoising and smoothing processing.Utilize convolutional neural network to combine multi-scale analysis to enhance fire point features, form feature set.Multiple satellite cooperative observation data are matched with feature set, and enhanced dataset of high frequency observation is generated by time series alignment.Adopt deep learning and transfer learning optimization model, extract key area features, share multiple satellite data through distributed computing, obtain coverage enhanced dataset.Based on timeliness evaluation model, dynamically adjust multiple satellite scheduling strategy, output scheduling result.Through reinforcement learning optimization algorithm robustness, combined with complex weather condition simulation, obtain robustness enhanced model, and utilize multi-source data verification technology to verify data fusion precision, output final monitoring result.The application significantly improves the accuracy, timeliness and coverage integrity of fire point detection.

The invention relates to a hopping beam scheduling method and a satellite constellation system. The method comprises the following steps: acquiring service cell information corresponding to a satellite, and performing feature extraction on the service cell information to obtain a service cell feature vector; receiving feature vectors of adjacent cells sent by other satellites adjacent to the satellite in the satellite constellation system, and fusing the feature vectors of the adjacent cells to obtain fused feature vectors of the adjacent cells; and based on the serving cell feature vector and the adjacent cell fusion feature vector, determining the number of hopping beams and cell identifiers of the satellite in each hopping beam time slot, and pointing the hopping beams matched with the number of the hopping beams to cells corresponding to the cell identifiers at the ending moment of the corresponding hopping beam time slot. By adopting the method, the effectiveness of each satellite scheduling hopping beam can be ensured, and the overall throughput of a satellite constellation system is improved.

The embodiment of the application provides a GNSS-R satellite constellation remote sensing data processing method and device based on a multi-thread technology, electronic equipment and a computer readable storage medium, and relates to the field of GNSS reflection signal remote sensing technology (GNSS-R). The method comprises the following steps: acquiring original data files of all satellites in a satellite constellation; scheduling a first number of computer threads to process the original data files to obtain a first data file corresponding to each satellite; scheduling a second number of computer threads for each satellite to process the corresponding first data file to obtain a second data file corresponding to each satellite; scheduling the first number of computer threads to process each second data file to obtain a third data file corresponding to each satellite; and scheduling a third number of threads for each satellite to process each third data file to obtain a target data file. The embodiment of the application can improve the timeliness of GNSS-R large-scale networking constellation data processing compatible with multiple GNSS systems.

An Earth monitoring system comprises a satellite constellation and a schedule calculation module located on Earth. The schedule calculation module may be configured to implement a satellite management method receiving a request post-launch for a location to be imaged; recalculating an existing schedule of the satellites to provide an updated schedule that includes imaging the location; and providing the updated schedule for transmission to one or more of the satellites.

The present invention provides a satellite mission planning method, system, storage medium, and electronic device based on a greedy adaptive annealing contract net algorithm, relating to the field of satellite scheduling. In the present invention, based on a preset slave satellite mathematical model, each slave satellite uses a greedy adaptive annealing contract net algorithm to generate a corresponding current bid proposal for a currently unscheduled task sequence. Based on a preset master satellite mathematical model, the master satellite uses a total observation benefit maximization evaluation strategy to evaluate bids, selecting a winning proposal from the bids, and ultimately outputting a globally optimal satellite mission planning proposal. By establishing a two-level decision-making mathematical model, the inconsistency between the primary objective and sub-objectives in distributed satellite mission allocation and scheduling problems, which is not considered in traditional contract nets, is resolved. A higher-quality proposal is obtained through multiple rounds of bidding and tendering.

This application discloses a regional target decomposition method for collaborative scheduling of satellite resources in heterogeneous remote sensing constellations, relating to the field of satellite remote sensing technology. The method includes: determining multiple stripes of satellite observation target areas in a heterogeneous remote sensing constellation; constructing triplet decision variables for each strip: strip start-point encoding, strip selection binary variable, and length normalization coefficient; constructing strip combination schemes based on the triplet decision variables; constructing a regional target decomposition model with the objective function of minimizing total satellite usage cost, combined with coverage constraints and strip-related constraints; solving the regional target decomposition model to obtain the optimal strip combination scheme; and obtaining the triplet decision variables for each strip based on the optimal strip combination scheme. This application provides reasonable input for subsequent single-satellite scheduling, helps generate better scheduling schemes, maximizes the fulfillment of mission requirements, fully utilizes the observation capabilities of heterogeneous remote sensing constellations, and ensures their efficient collaborative operation.

The invention relates to a GIS (Geographic Information System)-based agile satellite area target observation meta-task generation method, which comprises the following steps of: inputting sub-satellite point shp data of a satellite, filtering invisible points, and intersecting with shp data of an area to be observed to obtain an available satellite; generating preliminary visible windows according to continuous time grouping, and combining the windows according to satellite maneuvering time constraint; calculating the maximum visual distance of the satellite and generating a maximum visual area; extracting region boundary point data, and calculating a side swing angle range; dividing a to-be-observed area into grid center points, and generating a roll angle data set and candidate strips; and finally, optimizing and selecting an optimal strip and a side-sway angle through a task planning algorithm. According to the method, shp data is used as input, a geographic information system (GIS) technology is adopted to process a complex spatial relationship between a satellite visual range and a to-be-observed area, the spatial calculation efficiency and precision are remarkably improved, the solution space of a combined optimization problem is converted from infinity to finite, problem solving becomes possible, and the method is suitable for large-scale popularization and application. And a foundation is laid for satellite scheduling and planning scheme formulation.

The invention discloses a full-precision model scheduling method for a super-sensitive earth observation satellite, which comprises the following steps of: 1, accurately calculating attitude switching time of adjacent tasks through an Euler angle and a rotation matrix, 2, searching task insertion time based on attitude switching satisfaction deviation values of the adjacent tasks on two sides, and simultaneously realizing a local search algorithm for scheduling the super-sensitive satellite. Step 3, based on discrete integer coding and minimum graph decoding, realizing optimal path exchange and reserved crossover and mutation operators, and further realizing a global search evolutionary algorithm of hypersensitive satellite scheduling; step 4, based on a memetic algorithm combining the step 2 and the step 3, realizing scheduling high-efficiency solution of the full-precision model of the hypersensitive satellite; according to the method, the problem of efficient scheduling of the full-precision model of the hypersensitive satellite is solved.

The invention provides a rapid satellite visibility forecasting method based on a numerical orbit predictor. The method comprises the steps that on the basis of current orbit parameters of a satellite, a satellite orbit is forecasted with a first preset step length, and a first orbit forecasting point sequence is obtained; carrying out interpolation processing to obtain a preliminarily estimated continuous orbit; extracting a time period of which the elevation angle is greater than or equal to a preset elevation angle threshold as at least one preliminary visible time interval; carrying out integration according to a second preset step length to generate a corresponding second orbit forecast point sequence; respectively calculating the elevation angle of the satellite relative to the specified ground target based on each point in the second orbit forecast point sequence to obtain discrete elevation angle data; performing polynomial fitting on the elevation data to obtain an elevation interpolation polynomial; and subtracting the preset elevation angle threshold value to obtain a target elevation angle threshold value function, solving a zero crossing point of the target elevation angle threshold value function, and determining a starting moment of the satellite in the final visible time interval and an ending moment when the satellite leaves the visible state.

The invention provides a dynamic satellite reconnaissance system utilizing marine environment feedback, and relates to the field of electrical digital data processing, and the dynamic satellite reconnaissance system comprises an environment monitoring module, a data analysis module and a satellite scheduling module, the data analysis module is used for analyzing the collected environment data to predict an investigation area, and the satellite scheduling module is used for adjusting a satellite orbit and investigation parameters; according to the system, the monitored marine environment data is analyzed, the investigation task is dynamically adjusted, and the probability of investigating effective information is effectively improved.