Research On Mission Scheduling Technology Of Earth Observation Satellites

Earth observation satellites have the characteristics of wide coverage,long duration,no national boundaries,and time domain restrictions.They are widely used in military operations,such as regional operations,anti-terrorism,and joint exercises.As the core of satellite earth observation application,task scheduling has broad application prospects and research value.This thesis focuses on intensive or sudden task scheduling research.The current problems are:(1)the existing satellite mission scheduling is not complete enough;(2)the algorithm is not efficient;(3)the stability of the scheduling system is not strong.Therefore,in accordance with the various scheduling environment and strategies,this research aims to improve the utilization efficiency of the Earth observation satellite and maintain the stability of the scheduling system.The satellite mission synthesis and scheduling model under certain conditions and the satellite mission proactive scheduling and reactive scheduling model under uncertain conditions are constructed respectively,and different solving algorithms are designed according to different models.The main research work is as follows:(1)In this work,we analyze the working principle and workflow of the satellite,classify and summarize the satellite task scheduling scheme according to different factors.Then we elaborate the four basic elements of the satellite task scheduling model,including input,output,constraints and scheduling objectives.Furthermore,without affecting the scheduling model,the general characteristics are supposed generally.Finally,the basic process of solving the satellite task scheduling problem is given: scheduling preprocessing,model construction,algorithm solving and scheme verification.(2)There are often problems such as mutual exclusion of tasks and low benefit of the task for the intensive task scheduling.By synthesizing and analyzing the task,a satellite mission scheduling model based on task synthesis is constructed,which contains satellite resource constraints,inter-task attitude adjustment time constraints and task uniqueness constraints.The fireworks algorithm which includes the designed codec rules and fitness function is proposed to solve the model.The program has improved the benefit and efficiency of the satellite task scheduling under the determined conditions.(3)In view of the three types of uncertainties,such as satellite resource failure,cloud cover and emergency mission,considering the satellite resource failure probability,cloud coverage and primary and secondary window information as prior knowledge,a satellite task proactive scheduling model with maximum scheduling gain and maximum slack as the objective function is constructed.A discrete multi-objective particle swarm optimization algorithm is proposed to solve the model.The dynamic particle swarm updating method based on mutation and cross operation is designed.The external archive maintenance strategy based on dense distance deletion and the global optimal value selection strategy based on selection probability are proposed.The scheme can maintain the stability of the scheduling system and obtain large scheduling benefits,which effectively enhances the robustness of satellite mission scheduling under uncertain conditions.(4)Aiming at the uncertain factors in the scheduling process,an event-driven strategy based on triggering rules is designed.Considering four task changes,combined with satellite resource constraints and task time window constraints,a satellite task reactive scheduling model with maximum scheduling gain and minimized disturbance measure as the objective function is constructed.A heuristic algorithm considering task synthesis,insertion,shifting and replacement is proposed to solve the model.It can solve the scheduling scheme with high yield and small adjustment range in a short time,and effectively solve the satellite mission dynamics under uncertain conditions.Scheduling problem.