Research On Multi-target On-Board Cooperative Observation Technology For Agile Satellite

In the application of earth observation and space target detection,emergencies such as the spread of multiple fire points of forest fires and the outbreak of debris from space collisions will generate the need for rapid response observation of a large number of targets.Agile satellites have the characteristics of fast task response,high detection accuracy,and strong attitude maneuverability.The orderly scheduling of multiple agile satellites can meet the above observation requirements.At the same time,the multiobjective cooperative observation problem of agile satellites puts forward high timeliness and requirements for the dynamic interaction of inter-satellite information for the cooperative planning and scheduling of multi-satellite networking.In order to meet the requirements of real-time response of satellites in orbit,it is necessary to perform calculations based on satellite on-board computing capabilities algorithm design and optimization.This paper mainly takes the regional static multi-target problem in earth observation and the space dynamic multi-target cooperative observation problem in space target detection as the basic problems to carry out the research on multi-target on-orbit cooperative observation technology of agile satellites.The main research work and innovations of this article are as follows:(1)In view of the regional static multi-target cooperative observation problem,the analysis and modeling of multi-target observation tasks are carried out,and a heuristic forward-back chain optimization combination on-orbit multi-object observation sequence is innovatively proposed for single-satellite mission planning.The planning method realizes the planning of a large number of target observation paths in the application of hundreds of milliseconds aims at on-board application.(2)At the multi-satellite cooperation level,a constraint-based hierarchical collaborative observation strategy is designed.The strategy level allocates objectives through the parallel principle of capacity division,cost division and simple priority,and the planning layer uses chain optimization combination method to solve the observation sequence.The algorithm is compared with genetic algorithm and simulated annealing algorithm.The experimental results show that the method has the characteristics of low computational cost and high profit.(3)Aiming at the problem of cooperative observation of spatial dynamic multiobjects,a cooperative planning and scheduling process was designed,and the collaborative observation model modeling was completed according to key links such as stereo positioning,guided handover,and multi-object patrol,and system task planning and scheduling were carried out.The system task planning and scheduling are mathematically modeled and a collaborative task planning and scheduling solution framework based on the grouped hierarchical feedback mechanism is proposed,which is conducive to the real-time iterative optimization of system scheduling tasks.(4)Based on the quantitative analysis of the cooperative observation model of multi-target space dynamics,a dual coverage analysis based on the projection method and a global communication multi-constrained constellation design method are proposed,which reduces the complexity of constellation design.The network model of the low-orbit agile constellation was constructed,and the information flow analysis of cooperative observation tasks was carried out.Combining the analysis of information transmission requirements for different tasks,a collaborative task-oriented Delay optimization routing algorithm based on dynamic link load weighting is proposed.After simulation,this method meets the requirements of load balancing and high-efficiency transmission under the limited transmission capacity between satellites Constellation cooperative information network transmission requirements.(5)According to the structural symmetry and movement periodicity of the LEO agile constellation,multiple scheduling strategies are designed for grouping and layering at the global and regional levels,and the global regional control and duty grouping strategy and the dynamic fast grouping strategy based on relative motion analysis are proposed.In the task planning and observation window scheduling solution problem,a fast solution method combining the comprehensive multi-factors observation period evaluation method and the Priority,Max-Observation,Earliest Observation allocation principle is proposed.Through simulation analysis,with less computational cost the fast observation grouping and planning and scheduling scheme is solved.This paper analyzes the information acquisition,information processing,and information transmission links of the multi-target cooperative observation problem of the LEO agile constellation.Based on the quantitative model of observation and network,a framework for multi-satellite cooperation and resource scheduling is constructed,and a cooperative task-oriented approach is proposed.Based on the dynamic link load weighted delay optimization routing algorithm,comprehensive observation duration evaluation method and the principle of important and longest early allocation,and carried out global and regional management and control strategies for the temporal and spatial matching of tasks and resources.Through simulation analysis,the experimental results show that the task planning model and planning scheduling algorithm proposed in this paper are reasonable and effective.In the research,the onboard feasibility is fully considered.The algorithm design has the characteristics of low complexity and high profit,and the engineering application value is outstanding.It also enriched the theoretical research on the multi-objective collaborative observation problem.