Research On Collaborative Mission Planning Methods Of Autonomous Multi-satellite

With the continuous development of imaging satellite technology,the satellite's ability improves rapidly.At the same time,the demand of satellite users has become more and more complex.Many tasks have been unable to finish just by a single imaging of a single satellite,but need multiple satellites to perform multiple imaging process.At present,the traditional method of controlling satellites has many limitations,such as the complexity of the imaging process,the absence of satellite self-perception ability and so on.Based on the new challenges faced with satellite applications and the good develop opportunity for satellite technology,a new research comes into our study field,namely,multi-satellite autonomous mission planning technology.At present,China's research on autonomous satellites is still in its infancy.In the second chapter,this essay first studies the general description of multi-satellite cooperative task planning problem.After analyzes the needs of satellite control agency,this paper designs the goals of autonomous satellites collaborative mission planning system.And then,combined with the design of autonomous satellite's intellectual ability,this chapter establishes a framework of multi-satellite self-collaborative task planning process.In this framework,the intervention to satellites of ground system is weaken.With the knowledge of intelligent agent technology,this chapter also designs the multi-satellite autonomous collaborative planning system.The system is designed into two layers,which are multi-satellite cooperation layer and single autonomous satellite layer.In chapter three,the author elaborates the multi-satellite autonomous collaborative task planning system.Based on the Jade platform,the intelligent autonomous single satellite agent and the multi-satellite collaborative task management agent are constructed.The main structure of the intelligent autonomous single satellite agent contains the Blackboard,the Kernel Structure and the Agent Plugin Collection.The multi-satellite collaborative task management agent contains the Knowledge Library,Task Market and Agent Plugin Collection.This design makes the system has features of ‘distributed' and ‘plug-in',which make the system with good robustness and scalability.In the fourth chapter,this paper studies the specific multi-satellite autonomous collaborative task problem – geostationary orbit and low orbit multi-satellite cooperative task planning problem.After analyzes the core of the problem,this chapter establishes the task planning model for the low-orbit optical agile satellite and the geostationary satellite respectively,and designs the task planning algorithm.At the same time,a multi-satellite task assignment algorithm based on the idea of load balancing is designed.It is worth mentioning that the research of geostationary satellite mission planning algorithm is absent in China,and the HNSGA-II algorithm designed in this paper fills the blank.The fifth chapter is the simulation experiment,which mainly includes two parts.The first part is the experiment of geostationary satellite earth observation mission planning algorithm.The experiment is carried out for point targets with random distribution and the clustering feature distribution.The good performance of HNSGA-II algorithm solving this problem is proved.The second part is based on the multi-satellite cooperative task planning system designed in Chapter 3 of this paper.The experiment platform is carried out to verify the validity of the system designed in this paper.And a simulated collaborate mission scene is designed and finished successfully in this platform.