Research On Mission Planning Of Video Satellites For Ground Multi-object Staring Imaging

In recent years,video satellites have developed rapidly and have achieved fruitful results both at home and abroad.As a new type of earth observation satellites,they can perform staring imaging on a target area to obtain dynamic video information.In the entire video satellite imaging system,ground control and mission planning systems play an important role in the efficiency of imaging tasks.At the same time,as a typical agile satellite,the mission planning problem of video satellite is much more difficult than that of traditional imaging satellite.In this paper,the scheduling problem of video satellite for staring imgazing is discussed.According to the needs of different scenes,the model of the problem,algorithm design and application value are discussed and studied in order to further improve the imaging efficiency of video satellite in different scenes.This article mainly studies from the following aspects:Starting from the fundamental principle and working process of video satellite's staring imaging,the mission planning problem of staring imaging tasks is systematically described.The factors of the problem,the basic assumptions and the constraints involved in the mission planning are also discussed.The difficulty of this problem and its causes are analysised.A method to estimate the attitude maneuvering time is given,which provides the basis for modeling and solving of the mission planning problems.Aiming at the task planning problem of video satellite in emergency observation,this paper establishes a continuous time constraint satisfaction model for emergency task planning problem by analyzing the characteristics and the operating mechanism of staring imaging.Based on ant colony algorithm,an improved ant colony algorithm with tabu list(Tabu-ACO)is designed to solve this problem,and a satisfactory solution of the problem is obtained.Compared with the traditional time-discrete model and the complex network algorithm,the Tabu-ACO algorithm is proved to be much more effective in solving this problem.Based on the daily autonomic planning requirements of distributed imaging satellite system,the rapid solution of multi-satellite long-term mission planning is studied.After analyzing and modeling the multi-satellite long-term mission planning problem,the method of constructing solution and heuristic solution are used and a fast heuristic solution algorithm is designed.Finally,Simulation examples are given.The simulation results show that,when the number of observation tasks is less than 100,the Heuristic Fast Algorithm designed in this paper can get the scheduling results in 0.1 seconds.By comparing the computational speed of on-board computers and personal computers,the feasibility of on-board calculation of this algorithm is verified.Considering the possible changes of mission attributes during the task execution,the dynamic mission planning of video satellite is studied.Based on the characteristics of the problem and the two important planning goals,the dynamic mission planning model under the limitation of task change is proposed,which transforms the multiobjective problem into a single objective optimization problem with restricted conditions.And based on the model of "Exiting-Insert " operation,a heuristic algorithm for solving dynamic task planning is established.The result of the algorithm gives the different optimal mission revenues under different constraints of task change rate,and the significance of the result representation of the model built in this paper to the task planning decision-making is anylisied.