改进遗传与萤火虫算法的多星多目标任务规划

Multi-Satellite and Multi-Objective Mission Scheduling Based on Improved Genetic and Firefly Algorithms

针对航空航天领域中多卫星协同对地面多目标观测的任务规划问题,对实际任务场景进行建模仿真,并计算各卫星对各点目标的观测时间窗。提出了一种考虑到卫星观测所受多种约束的改进的遗传算法,用于解决传统遗传算法无法对不可行解做单独处理的问题,同时,进一步使用萤火虫算法对改进遗传算法进行优化,利用萤火虫算法良好的局部优化能力对遗传算法的后期寻优性能进行改善。最终将这一组合算法与贪心算法、单独的改进遗传算法以及遗传-退火算法进行实验对比,证明了改进遗传-萤火虫算法的性能更加优越。

Aiming at the mission planning problem of multi satellite cooperative observation for ground multitarget in aerospace,we simulated the actual mission scene and calculated the observation time window of each satellite for each target.On this base,an improved genetic algorithm considering various constraints of satellite observation was proposed to solve the problem that the traditional genetic algorithm cannot deal with the infeasible solution separately.At the same time,firefly algorithm was further used to optimize the improved genetic algorithm,and the good locality of firefly algorithm was used to improve the later optimization performance of genetic algorithm.Finally,this combination algorithm was compared with greedy algorithm,improved genetic algorithm and genetic annealing algorithm,which has proven that the performance of our improved genetic and firefly algorithm is better.