基于改进遗传算法的多无人机协同侦察航迹规划

Track planning of multi-UAV cooperative reconnaissance based on improved genetic algorithm

针对多无人机在复杂的战场环境中如何高效侦察多种类型目标的问题,提出了一种基于改进遗传算法的多无人机协同侦察航迹规划算法。首先,根据战场环境中不同类型目标的侦察要求,建立了以整体时间代价最小为目标函数的航迹规划模型。然后,改进了遗传算法的编码、交叉和变异等操作,来实现异质型目标分配与航迹规划的集中一体化求解。最后,为了提高算法的收敛速度,在交叉和变异操作过程中,加入了适应度更新策略。仿真结果验证了该算法的可行性,而且在4架无人机观测3种类型共10个目标的仿真条件下,相较于使用双染色体编码和多重变异算子的基于对立的遗传算法,时间代价降低了5.79%,收敛速度提高近12倍,有效提高了航迹规划的精度及效率。

Aiming at the problem of how to efficiently reconnoitre multiple types of targets for multiple unmanned aerial vehicles(multi-UAV)in complex battlefield environments,a multi-UAV cooperative reconnaissance trajectory planning algorithm based on improved genetic algorithm is proposed.Firstly,according to the reconnaissance requirements of different types of targets in the battlefield environment,the track planning model with the minimum overall time cost as the target function is established.Then,the coding,crossover and mutation of the genetic algorithm are improved to realize the centralized and integrated solution of heterogeneous target assignment and track planning.Finally,in order to improve the convergence speed of the algorithm,the fitness update strategy is added in the process of crossover and mutation operation.The simulation results verify the feasibility of the algorithm.Moreover,compared with the opposition-based genetic algorithm using double-chromosomes encoding and multiple mutation operators(OGA-DEMMO),under the simulation conditions where 4 drones observe 3 types of 10 targets,the time cost of the proposed algorithm is reduced by 5.79%and the convergence speed is increased by nearly 12 times,which effectively improves the accuracy and efficiency of trajectory planning.