无人机三维路径规划的粒子群混合算法

A Hybrid Particle Swarm Optimization Algorithm for 3D Path Planning of UAV

路径规划是无人机控制过程中的重要环节之一,现有基于粒子群等算法的传统路径规划方法存在容易陷入局部最优等问题,无法适应现实场景中复杂环境及高搜索速度的要求。针对已有方法的缺陷,提出了一种无人机路径规划的高性能细菌觅食-遗传-粒子群混合算法,以传统粒子群优化算法为基础,引入细菌觅食算法及遗传算法思想,提高算法计算速度与能力,同时考虑实际场景中无人机的运行约束,进一步提高了方法的可用性。最后,利用仿真实验验证了所提方法的有效性,并通过与传统方法对比证明了所提方法在运行时间、规划航程等方面的优越性。

Path planning is one of the most important parts in unmanned aerial vehicle(UAV)controlling.The existing methods of path planning for UAV using traditional optimization algorithms have difficulties in dealing with complex planning problems because of the poor calculation ability and low speed.For the shortcomings of existing methods,a hybrid algorithm for UAV path planning based on traditional particle swarm optimization(PSO)is proposed,in which bacterial foraging optimization(BEO)and genetic algorithm(GA)are introduced to improve the searching ability.Besides,the constraints of environment in real scenario are considered.The effectiveness of the proposed method is verified by simulation.Comparison of several algorithms shows the advantages of the hybrid algorithm in calculation ability and performance.