Orbit Design And Task Planning Of Optics Surveillance Satellite For Space Objects

As the demands of space objects surveillance move to smaller size,more detailed and huge number,the surveillance ability and catalogue level should be significantly improved.Provided with the advantages of both space-based platform and optics sensor,the space-based optics surveillance has become a crucial method and studied by researchers from all over the world.For the purpose of improving the observation efficiency of space-based optics surveillance system,this dissertation studies the orbit optimization problem,the task merging problem and the task planning problem of space-based optics surveillance satellites.The main results achieved in this dissertation are summarized as follows.The space-based optics surveillance constraint models and solving algorithms are improved and the surveillance performance criterions are defined.1)Four optics visibility constraint models including the Earth masking,the Earth shadow,the sunlight and the moonlight are established.2)Four observability constraint models including the field of view,the observable star magnitude of space object,the relative angular velocity and the background of view are proposed.3)For the instance of large difference between every single satisfaction degree,a multi-satisfaction algorithm based on adaptive sorting method is proposed,and the calculation was cut down by 40% compared to the basic method.The orbit optimization approaches of remote detection are studied and proposed.1)Five performance criterions of remote detection including arc length,revisit interval,regressive period,observability and coverage ratio are established.2)Considering the detection performance difference of remote relative movement,four relative orbit elements including relative inclination,relative RAAN,relative earth angle and relative distance are defined,and the disciplinarian of relative orbit elements refer to orbit element bias between surveillance satellite and space object is simulated.3)An optimization algorithm merged by random sampling method and multi-mutation adaptive Genetic Algorithm is proposed to solve orbit design of surveillance satellite for space objects with large sample size,and the computation efficiency is greatly improved without quality losses.The orbit optimization approaches of reconnaissance imaging surveillance are studied and proposed.1)Three performance criterions of reconnaissance imaging surveillance including arc length,minimum distance and relative angle rate are established.The typical configuration of small bias relative motion in tangent plane and the influence caused by eccentricity bias and semi-major bias are analyzed.2)The analytic solutions of relative ascending node reconnaissance problem is derivated in case of two bodies,and three reconnaissance strategies including orbit control reconnaissance mode in relative ascending node,drift reconnaissance mode and mix reconnaissance mode.3)According to the clustering of space objects' orbit,a packing pruning method is proposed to acquire the global optimal solution in limited time,while a segment pruning method is used to reduce the sensibility of orbit plane drift.The planning approaches of space object surveillance are studied and proposed.1)The space object surveillance model is set up according to the constraint of storage capacity and power capacity;2)A task merging method using particle swarm optimization is proposed to improve the profit of imaging.3)Two panning algorithms including tabu search algorithm and knowledge-based ant colony optimization are proposed to solve the imaging and data transmission planning problem with better efficiency compared to 0-1 integer programing.