融合随机反向学习蜜獾算法的无人机三维路径规划

UAV 3D Path Planning Based on Random Opposition-based Learning Honey Badger Algorithm

提出一种融合随机反向学习的蜜獾算法(random opposition-based learning honey badger algorithm,ROBLHBA)以解决复杂的无人机三维路径规划问题。在原算法的挖掘阶段和采蜜阶段后加入随机反向学习策略,增强算法的全局能力,加快了收敛速度。此外,为了更加真实地模拟复杂的无人机三维路径规划问题,将路径长度和与威胁障碍的碰撞情况作为代价函数,将无人机三维路径规划问题转化为对无人机的安全性和飞行操纵约束下的优化问题,使仿真实验更加具有真实性。实验对比多种群智能优化算法在同一场景下的运行结果,同时也将改进算法在不同场景下运行。实验结果体现改进后的蜜獾算法在无人机三维路径规划问题上有更好的实用性和鲁棒性。

This paper presents a random opposition based learning honey badger algorithm(ROBLHBA)to solve the complex three-dimensional path planning problem of UAV.The random reverse learning strategy is added to the original algorithm after the mining stage and the honey collecting stage,which enhances the global ability of the algorithm and accelerates the convergence speed.In addition,in order to more truly simulate the complex UAV three-dimensional path planning problem,this paper takes the path length and the collision situation of threat obstacle as the cost function,and transforms the UAV three-dimensional path planning problem into an optimization problem under the constraints of UAV safety and flight control,so that the simulation experiment is more realistic.The experiment compares the running results of several swarm intelligence optimization algorithms in the same scenario,and the improved algorithm also runs in different scenarios.The experimental results show that the improved honeybadger algorithm has better practicability and robustness in UAV 3D path planning.