Research On Multi-type Task Combination Scheduling Method Of Optical Imaging Agile Satellite

With the development of space technology,a new generation of agile optical imaging satellites gradually developed,which has three-axis space attitude maneuverability,greatly expand the satellite's mission capability,suitable for military reconnaissance,disaster observation,weather Monitoring and other fields.Agile satellites,due to their mobility and flexibility,can use observing patterns that are different from traditional satellites along non-track-based push-ups in the observation of the target,and thus have a more flexible observation method when dealing with different types of tasks.In this paper,the task scheduling method of agile satellites in the face of multiple types of task combinations is studied.Firstly,the agile satellite mission scheduling model is established.Combining the characteristics and constraints of different tasks,a constraint satisfaction model for satellite-to-ground observation is established.Aiming at the maneuverability of agile,the method of calculating the attitude and the time of gesture conversion between the missions is designed.Aiming at the orbital motion of the satellite and the rotation motion of the earth,the satellite panning imaging model is established,and the calculation method of the observation time and observation window is deduced.Then,for different types of tasks,this paper defines the strategy of strip division,analyzes the starting point of strip division,and determines the different types of tasks to be transformed into strip tasks as a solution to the task scheduling problem.Then,the band partitioning algorithm is designed for the dense point target,the regional target,the curve target and the corresponding task characteristic and constraint condition respectively.In this paper,a band generation algorithm based on group partition is designed to generate all the strip nodes formed by the combination of point objects satisfying the constraint of strip width.For the regional target,this paper jumps out of the traditional parallel stripe division strategy,designed a multi-angle combination of strip division strategy.This band division strategy is more adaptable to the agility of agile satellites and improves the efficiency of observation.For the curve target,this paper designs an ordered search strip generation strategy,divides the curve target For a number of continuous strip,reducing the satellite attitude control pressure,improve the observation efficiency.Finally,for the result of the strip division,this paper establishes the meta-task model,integrates the generated strip task and the point target task into meta-tasks,and forms the meta-task library to facilitate the scheduling algorithm.At the same time,considering the complexity of the problem,this paper designs two algorithms to complete the scheduling task.The meta-task selection algorithm based on genetic algorithm is designed as the upper layer algorithm to complete the selection of meta-tasks.The genetic algorithm is combined with the content of this paper,and the corresponding coding method and selection operator,crossover operator and mutation operator are designed.The genetic algorithm is combined with the research content of the genetic algorithm.Genetic operators,so that the task of choice to solve the task.This paper designs a heuristic task sorting algorithm based on tabu search to complete the task sorting,designs the heuristic algorithm based on the initial solution generation algorithm,designs the four neighborhoods,designs the tabernment list and the amnesty criterion,and makes the algorithm better The completion of the observation sequence to complete the task.Finally,a representative simulation example is designed,which verifies the rationality and validity of the proposed algorithm from many perspective.