English

中国农机化学报

中国农机化学报 ›› 2024, Vol. 45 ›› Issue (9): 234-243.DOI: 10.13733/j.jcam.issn.2095-5553.2024.09.036

• 农业智能化研究 • 上一篇    下一篇

基于混合改进麻雀搜索算法的农用移动机器人路径规划

牟远明 1,卓然 1,高飞 2   

  1. (1.浙江特殊教育职业学院,杭州市,310023; 2.浙江工业大学计算机科学与技术学院,杭州市,310000)
  • 出版日期:2024-09-15 发布日期:2024-09-04
  • 基金资助:
    浙江省自然科学基金探索项目(LTGG23F020003)

Path planning of agricultural mobile robot based on hybrid improved sparrow search algorithm 

Mou Yuanming1,Zhuo Ran1,Gao Fei2   

  1. (1. Zhejiang Vocational College of Special Education,Hangzhou,310023,China; 2. School of Computer Science and Technology,Zhejiang University of Technology,Hangzhou,310000,China) 
  • Online:2024-09-15 Published:2024-09-04

摘要:

针对目前农用移动机器人进行全局路径规划容易产生局部最优路径的不足,提出混合多策略改进麻雀搜索算法的移动机器人路径规划算法。首先,为提升麻雀搜索算法的寻优能力,引入偶对称无限折叠混沌序列、螺旋式搜索发现者更新、多重学习追随者更新机制对传统算法的种群多样性、搜索盲目性及全局搜索能力改进,实现多策略混合改进麻雀搜索算法 MHISSA。然后,构建移动机器人的路径规划模型,利用导航点定义种群个体的编码方式,并构造同步避障与路径最短的适应度函数,结合三次样条插值法将 MHISSA应用于求解农用移动机器人全局路径规划问题。最后,构造简单和复杂障碍物环境对单机器人路径规划和多机器人协同路径规划进行试验对比分析。结果表明,改进算法能够得到平滑无碰撞的最优路径,路径规划长度最优值和均值较传统麻雀搜索算法平均分别降低 23. 08%和 19. 56%。实地场景案例也证明方法在农用移动机器人路径规划方面具备较好的实用性

关键词: 农用机器人, 路径规划, 麻雀搜索算法, 螺旋搜索, 多重学习

Abstract:

Aiming at the shortcomings of current agricultural mobile robot path planning methods that are prone to produce a local optimal path,a mobile robot path planning algorithm based on hybrid multi-strategy improved sparrow search algorithm is proposed. Firstly,for improving the optimizing ability of sparrow search algorithm,the even symmetric infinite folding chaotic sequence,spiral search finder update,multiple learning follower update are introduced to improve the population diversity,search blindness and global search ability of the traditional algorithm,so as to realize the Multi-Strategy Hybrid Improved Sparrow Search Algorithm(MHISSA). Then,the path planning model of mobile robot is constructed,and the coding method of population individuals is defined by using the navigation point model,and the fitness function of synchronous obstacle avoidance and shortest path is constructed. Combined with cubic spline interpolation,MHISSA algorithm is applied to solve the global path planning problem of mobile robot. Simple and complex obstacle environments are constructed,and the experimental comparative analysis of single robot path planning and multi-robot cooperative path planning are carried out. The results show that the improved algorithm can get a smooth and collision-free optimal path,the optimal value and the average value of path planning length are 23. 08% and 19. 56% lower than those of the traditional sparrow search algorithm respectively. The field scene case verification proves that the method has good practicability in the field of path planning for agricultural mobile robots.

Key words: agricultural robot, path planning, sparrow search algorithm, spiral search, multiple learning

中图分类号: