| The autonomous mission planning process of agile earth observation imaging satellite can not rely on ground computing,and the mission objectives can be planned by the satellite itself,which saves a lot of ground resources,reduces the transmission pressure and reduces a lot of satellite waiting time.Therefore,the research of agile satellite autonomous mission planning has high value.Due to the limited satellite computing resources,the computational efficiency of agile satellite autonomous task processing and autonomous task planning method is higher than that of ground task processing and task planning method;In the face of changing environment,agile satellite autonomous task planning method needs to have the ability of rapid response to changing tasks,and the planning results have the ability of global planning.Taking the agile satellite autonomous task planning algorithm as the research object,this paper deeply studies the agile satellite autonomous task processing process,mainly including the calculation of visible time window and the dynamic decomposition of complex region;The planning efficiency of agile satellite autonomous planning algorithm is optimized;The autonomous task decision algorithm is added to improve the dynamic response ability of the satellite on the basis of taking into account the global planning ability;The effectiveness of each research is verified by simulation.Firstly,the autonomous task processing process of agile satellite is designed,and a visible time window calculation method for regional targets is proposed.The geometric judgment conditions for the visibility of agile satellite to point targets are analyzed,and the calculation method of rough search and accurate calculation is designed.In order to make this method applicable to point targets and regional targets,in addition to searching and calculating the satellite ephemeris,the boundary point finding method is designed to search and calculate the arc boundary,Get the visible time window containing the arc area.The simulation results show that there is only 0.1s error in the time window calculation and STK calculation results of the experimental method,and the calculation efficiency is 31.2%higher than that of the traditional dichotomy method;There is only 0.1s error between the time window calculation and STK calculation results of regional targets,and the calculation efficiency is 29.3%higher than that of geometric analysis method.Secondly,a dynamic decomposition method based on complex regional targets is also carried out in the autonomous task processing process of agile satellite.A dynamic decomposition algorithm of regional targets is designed to decompose a complex regional target into multiple strip targets that can be observed at one time.In this scheme,the attitude angle is taken as the optimization target,and the strip is divided into equal width strip and unequal width strip.The influence of the two strip decomposition methods and non trace strip decomposition methods on the imaging coverage,imaging time and imaging accuracy of regional targets.The simulation results show that the area coverage of unequal width strip is 9.8%higher than that of equal width strip,and the imaging accuracy of unequal width strip is 23.5%higher than that of equal width strip;Compared with the non trace decomposition method,the unequal width strip decomposition method is higher than the non trace strip decomposition method in coverage and imaging accuracy by 0.03 and 0.09 respectively.The average time of the unequal width strip decomposition method is 5.2s less than the non trace strip decomposition method.The unequal width strip decomposition method has better imaging coverage,imaging accuracy and decomposition efficiency.Then,in view of the limited satellite computing resources and how to reasonably allocate multiple observation tasks to their observation time faster than the ground and current planning methods,this part designs the construction and solution algorithm of agile satellite autonomous task planning model.the input and output information of the model is determined,and the constraints to be met are analyzed.the general algorithm framework is selected to appropriately cut the adaptive large neighborhood search algorithm according to the problem model,and seven sorting rules for generating initial solutions are added to improve the ability of the algorithm to find better initial solutions.When updating the weights of neighborhood operators,the inner loop structure is added to avoid the randomness of operators,and the simulated degradation algorithm is used as the acceptance criterion,By improving the global search ability of the algorithm,an agile Autonomous Task Planning Algorithm with small storage space,simple program and fast solution speed is designed to determine the reasonable imaging target sequence.compared with the research results of agile satellite autonomous planning algorithm and ground satellite planning algorithm,the optimization results of the algorithm are analyzed,which proves that the optimization income of the algorithm is higher and the calculation time is less.Finally,aiming at the problem that the agile satellite autonomous task planning algorithm can not take into account the dynamic response ability and global planning ability when planning the changing tasks,the agile Satellite Autonomous Task decision-making process is proposed.Make decision-making strategies,determine the planned task scale and planning cycle,and solve the problems of which tasks need to be planned and when to plan in the process of agile satellite autonomous task planning.An agile satellite autonomous task decision algorithm based on multi-objective hybrid genetic ant colony algorithm is proposed.The task scale and planning cycle of agile satellite planning are taken as the output,and the task revenue,maneuver angle and disturbance to the initial scheme are taken as the planning objectives for modeling.Combined with the autonomous task planning process,the average planning completion rate of dynamic high-yield tasks is 92.3%,which is much higher than 31.2%of rolling planning decision-making and 67.1%of sequential decision-making.The Autonomous Task decision-making scheme has better dynamic response ability and global planning ability than the existing decision-making scheme. |
PDF Full Text Request