English

中国农机化学报

中国农机化学报 ›› 2025, Vol. 46 ›› Issue (6): 142-149.DOI: 10.13733/j.jcam.issn.2095-5553.2025.06.021

• 农业智能化研究 • 上一篇    下一篇

基于A*与DWA算法融合的割草机器人路径规划

刘士琪1,闫九祥1,李倩1,温以民2,刘强2   

  1. (1. 齐鲁工业大学(山东省科学院)山东省科学院自动化研究所,山东省机器人与制造自动化技术重点实验室,济南市,250014; 2. 临沂市金立机械有限公司,山东临沂,276000)

  • 出版日期:2025-06-15 发布日期:2025-05-22
  • 基金资助:
    山东省自然科学基金(ZR2023MF077);中央引导地方科技发展专项资金(YDZX2022086)

 Path planning of mowing robot based on the fusion of A* and DWA algorithm

Liu Shiqi1, Yan Jiuxiang1, Li Qian1, Wen Yimin2, Liu Qiang2   

  1. (1. Institute of Automation, Qilu University of Technology (Shandong Academy of Sciences), Shandong Provincial Key Laboratory of Robot and Manufacturing Automation Technology, Jinan, 250014, China; 2. Linyi Jinli Machinery Co., Ltd., Linyi, 276000, China)

  • Online:2025-06-15 Published:2025-05-22

摘要:

针对室外环境下智能割草机器人割草作业路径规划及实时避障问题,建立割草机器人运动学数学模型,提出一种基于动态窗口法(DWA)和A*算法融合的割草机器人路径规划算法。基于A*算法实现割草作业地图的全局路径规划,基于动态窗口法构建全局最优路径的评价函数,通过激光雷达进行局部障碍物识别,实现局部路径的动态避障。仿真结果表明,在面对多个障碍物和多个目标点时,割草机器人的运行轨迹平滑,系统保持良好的稳定性。在搭建的真实环境下进行试验,割草机器人可实现全局路径自主导航及局部避障。试验过程中,割草机器人最大线速度为0.98 m/s,平均线速度为0.243 m/s,均在机械设计范围内,路径跟踪误差小于0.24 m,定位误差小于0.07 m,满足实际作业需求。

关键词: 割草机器人, 避障, 路径规划, 动态窗口法, A*算法

Abstract:

Aiming at the problem of path planning and real‑time obstacles avoidance of intelligent mowing robot in outdoor environments, a kinematic mathematical model of mowing robot is established, and a path planning algorithm for mowing robot based on the fusion of dynamic window approach (DWA) and A* algorithm is proposed. The global path planning for mowing operation maps is implemented based on the A* algorithm. The evaluation function for the global optimal path is constructed based on the dynamic window approach. The local obstacles recognition is performed by using the LiDAR to achieve dynamic obstacles avoidance of local paths. The simulation results show that when facing multiple obstacles and target points, the running trajectory of the mowing robot is smooth and the system maintains good stability. Experiments are conducted in a real environment, and the mowing robot can achieve global path autonomous navigation and local obstacle avoidance functions. During the experiment, the maximum linear velocity of the mowing robot was 0.98 m/s, and the average linear velocity was 0.243 m/s, both of which were within the mechanical design range. The path tracking error was within 0.24 m, and the positioning error was less than 0.07 m, meeting the practical operational requirements.

Key words: mowing robot, obstacle avoidance, path planning, dynamic window approach, A* algorithm

中图分类号: