基于多策略改进北方苍鹰算法的无人机路径规划

UAV Path Planning Based on Multi-Strategy Improvement of Northern Goshawk Algorithm

为实现在复杂地域和障碍环境下无人机高效、安全飞行路径规划问题,以飞行距离、转向角度、飞行高度为性能指标,以地形、空域中的障碍为约束建立了无人机三维路径规划模型.智能算法在解决高维多约束情况下无人机路径规划问题时,存在初始种群多样性差且分布不均、容易陷入局部最优解等局限性,对此提出了一种多策略改进的北方苍鹰算法(MINGO).该算法在初始化时借助Piecewise混沌映射有效地增强了初始种群的分布均匀性,借助Levy游走和随机扰动策略增强了种群的多样性和个体的质量,使算法快速且高精度的收敛,使用非线性收敛因子策略来有效平衡算法的全局和局部搜索能力,通过精英个体引导策略有效提升了算法的寻优效率.比较性的数值实验和仿真实验表明,MINGO具有更快的收敛速度和更高的寻优精度,处理复杂障碍环境下无人机三维路径规划问题具有非常明显的优势.

To tackle the efficient and safe path planning challenge for unmanned aerial vehicles(UAVs)in complex terrains and obstacle-rich environments,a three-dimensional path planning model is developed.This model considers flight distance,turning angle,and flight altitude as performance indicators while incorporating terrain and airspace obstacles as constraints.In response to limitations faced by intelligent algorithms in solving high-dimensional,multi-constraint UAV path planning problems,such as poor initial population diversity,uneven distribution,and susceptibility to local optima,a multi-strategy improved Northern Goshawk Algorithm(MINGO)is proposed.The algorithm effectively enhances the uniformity of the initial population distribution by leveraging Piecewise Discrete Mapping during initialization.Additionally,it enhances the diversity of the population and the quality of individuals by incorporating Levy flights and random perturbation strategies.This ensures rapid and high-precision convergence of the algorithm.A non-linear convergence factor strategy is employed to effectively balance the global and local search capabilities of the algorithm.Furthermore,the algorithm's optimization efficiency is effectively enhanced through an elite individual-guided strategy.Comparative numerical and simulation experiments demonstrate that MINGO exhibits faster convergence speed and higher optimization accuracy.It demonstrates significant advantages in addressing the three-dimensional path planning problem for UAVs in complex obstacle-rich environments.