汪杨凯，曾宏宇，赵然，许悦，张勇

摘 要：针对无人机的全局航迹规划进行专项研究，在分析了细菌觅食算法和粒子群算法的优劣势以后，建立了一种引用细菌觅食算法的趋化及迁徙算子的改进型粒子群算法，用以改进无人机的全局航迹规划问题。围绕研究主题，确定了以下技术路线，首先分析了细菌觅食算法、粒子群算法的实现过程，然后剖析了粒子群算法的缺点，提出了一种应用细菌觅食算法的趋化及迁徙算子的新型粒子群算法。面向无人机航迹规划的需求，分析了三维粒子群航迹规划模型、适应度函数、航迹平滑方法、算法早熟收敛判断等，最后利用Matlab软件进行仿真分析。通过与传统粒子群算法作对比，验证了改进后算法在稳定性和寻优能力方面具有明显优势。

关键词：航线自主;粒子群算法;无人机巡检;变电站

中图分类号：TM63 文献标识码：A     文章编号：1001-5922（2021）12-0173-05

Inspection of substation UAV based on autonomous route planning

Wang Yangkai， Zeng Hongyu， Zhao Ran， Xu Yue， Zhang Yong

（State Grid Hubei Electric Power Co.， Ltd.， Maintenance Company， Wuhan 443000， China）

Abstract：The global route planning of UAV is studied. After analyzing the advantages and disadvantages of bacterial foraging algorithm and particle swarm algorithm， a improved particle swarm algorithm using chemotaxis and migration operator of bacterial foraging algorithm is established to improve the global route planning of UAV. The following technical routes are determined： firstly， the implementation process of bacterial foraging algorithm and particle swarm algorithm is analyzed; and then the shortcomings of the particle swarm algorithm are analyzed; thus a new particle swarm algorithm using chemotaxis and migration operators of the bacterial foraging algorithm is proposed. For the requirements of UAV route planning， the three-dimensional particle swarm route planning model， fitness function， route smoothing method， algorithm precocious convergence judgment， etc. are analyzed， and finally the Matlab software is used for simulation analysis. By comparing with the traditional particle swarm algorithm， it is verified that the improved algorithm has obvious advantages in stability and optimization ability.

Key words：route autonomy; particle swarm optimization; UAV inspection; substation

0 引言

全局航跡规划指的是综合分析飞行环境、飞行约束、突发威胁等信息要素，最终确定一条确保飞行安全、达成飞行任务的全局航迹。全局航迹规划的前提条件就是快速解析各类信息，由于运算体量极大，常规的机载CPU难以应付，所以必须在无人机起飞前即利用地面计算机做好航迹规划，引导无人机在全域内进行安全飞行。这一策略的实时性不佳，却能够保障航迹规划的精度。本文提出的改进粒子群算法是以原始粒子群算法为基础，通过引入细菌觅食算法的趋化及迁徙算子，用以规避原始粒子群算法的缺点，由此构建基于改进粒子群算法的三维航迹规划模型，然后利用B样条曲线法对航迹进行平滑化处理，最终完成离线全局航迹规划。……