一种面向城市战的无人机路径规划群智能算法

A Swarm Intelligence Algorithm for UAV Path Planning in Urban Warfare

针对传统路径规划算法用于解决无人机战术路径规划问题时存在局部最优、慢收敛等问题,提出一种改进的群智能算法,以改变初始种群的更新方式,提升灰狼算法的全局寻优能力并加快其收敛速度。引入莱维飞行随机策略以及共生生物搜索算法,借助莱维飞行策略更新种群个体,利用共生生物搜索算法偏利共生阶段的交互性避免陷入局部最优问题。威胁建模是UAV路径规划的重要前提,对威胁物进行等效预处理,结合适应度函数检验算法。设计一种虚实映射仿真验证平台,通过以实映虚的手段验证算法的有效性。实验结果表明,改进后的算法改善了传统路径规划算法的局部最优和慢收敛等路径规划问题,对UAV作战能力的提升具有一定参考价值。

For the traditional path planning algorithm used to solve the tactical path planning problem of unmanned aerial vehicle,there are some problems,such as local optimum and slow convergence.In this paper,an improved grey wolf algorithm is proposed,which changes the updating mode of the initial population,improves the global optimization ability of grey wolf algorithm and speeds up its convergence speed.The Lévy flight random strategy and symbiotic search algorithm are introduced.The Lévy flight strategy isused to update the population individuals,and the symbiotic search algorithm is usedto avoid the local optimal problem.Threat modeling is an important prerequisite for UAV path planning,and the equivalent preprocessing of the threats is performed,and the algorithm is verifiedby combining the fitness function.And a simulation verification platform,incorporating virtual-to-real mapping,was designed to validate the algorithm s effectiveness through a real-to-virtual approach.The experimental results show that the improved algorithm effectively improves the path planning problems such as local optimum and slow convergence of the traditional path planning algorithm,and has a certain reference value for the improvement of UAV combat capability.