Autonomous Mission Planning For Agile Imaging Satellite

Due to its strong attitude maneuverability,agile imaging satellite has more flexibility and higher efficiency in the imaging task.With the increasing complexity of earth observation tasks,the real-time management and control burden of the ground measurement and control center on the satellite is increasing,so it is urgent for the satellite to have certain independent planning and decision-making ability to improve the efficiency of task execution.Satellite autonomous mission planning technology is an important technology for satellite to achieve autonomy and intelligence.The task planning problem of agile imaging satellite has many constraints and wide search space.It has been proved to be a NP hard problem.In this paper,we mainly focus on the task pre planning and re planning of single agile imaging satellite for multi-point targets,and form the following research results:Aiming at the problem of autonomous mission pre planning of imaging satellite,the constraints of target time window,satellite storage capacity,satellite attitude maneuverability,satellite orbit and earth rotation are considered.The target observation sequence,target start observation time and satellite are taken into account to maximize the mission observation revenue and minimize the energy consumption The pre planning model of imaging satellite mission is established by setting the attitude trajectory as the optimization variable.A solution framework based on adaptive genetic algorithm and Gauss pseudospectral method is designed,which makes full use of the advantages of the two methods.The solution of observation sequence and starting observation time is completed by using adaptive genetic algorithm.Then the satellite attitude direction of each time point is obtained according to the solution results.Finally,the optimal attitude maneuver path is solved by using Gauss pseudospectral method.In the adaptive genetic algorithm,a onedimensional real coding and decoding method based on the relative start observation time is designed to complete the coding of the observation sequence and the start observation time two variables to be optimized.This coding method can ensure that the observation sequence that does not meet the constraints will not be generated in the process of crossover and genetic,which greatly improves the iteration efficiency.At the same time,the coding method can also be used to solve the problem when the target point has multiple time windows.Aiming at the task re-planning problem of agile imaging satellite,this paper considers it as a kind of dynamic insertion problem of new tasks,analyzes several new task insertion scenarios,and designs three-level progressive heuristic task sequence solution methods for each scenario: task direct insertion,task local adjustment and task deletion.At the same time,an iterative method to determine the optimal observation time of the target point is designed to determine the start observation time of the inserted target point and its adjacent tasks.Simulation results show the effectiveness of the proposed heuristic algorithm.