Scholarlink

<p>The study proposes a fusion improvement algorithm aimed at avoiding dynamic obstacles such as animals or running agricultural machinery from interfering with the movement of mobile robots towards target points in agricultural environments (path planning). It chooses Grey Wolf Optimization (GWO) algorithm for improvement. Firstly, the population initialization method and convergence factor of GWO were improved, and GWO was combined with Particle Swarm Optimization (PSO) algorithm. The improved Grey Wolf Optimization (IGWO) algorithm was compared with various algorithms, and its performance was improved by 11.5%, 14.4%, and 1.6% compared to Ant Colony Optimization (ACO), PSO, and GWO, respectively. The IGWO has reduced 55.6% of turns compared to GWO resulting in a smoother path. Then, combine IGWO with Dynamic Window Approach (DWA). The key path points of the IGWO algorithm are used as intermediate target points for the DWA algorithm. Finally, the motion trajectory of the robot was simulated and a globally optimal path that adapts to the dynamic environment based on key information was generated. The simulation results demonstrate that the mobile robot can avoid all dynamic obstacles and the proposed method is effective.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Dynamic path planning of mobile robots based on improved GWO algorithm

  • 期刊论文

  • 发表日期：

    2025年6月19日

查看全文
  • Ximing Zhu,
  • Jianpeng Zhao,
  • Hengxin Ni

摘要

The study proposes a fusion improvement algorithm aimed at avoiding dynamic obstacles such as animals or running agricultural machinery from interfering with the movement of mobile robots towards target points in agricultural environments (path planning). It chooses Grey Wolf Optimization (GWO) algorithm for improvement. Firstly, the population initialization method and convergence factor of GWO were improved, and GWO was combined with Particle Swarm Optimization (PSO) algorithm. The improved Grey Wolf Optimization (IGWO) algorithm was compared with various algorithms, and its performance was improved by 11.5%, 14.4%, and 1.6% compared to Ant Colony Optimization (ACO), PSO, and GWO, respectively. The IGWO has reduced 55.6% of turns compared to GWO resulting in a smoother path. Then, combine IGWO with Dynamic Window Approach (DWA). The key path points of the IGWO algorithm are used as intermediate target points for the DWA algorithm. Finally, the motion trajectory of the robot was simulated and a globally optimal path that adapts to the dynamic environment based on key information was generated. The simulation results demonstrate that the mobile robot can avoid all dynamic obstacles and the proposed method is effective.