基于动态自确定并行模糊聚类鸡群优化算法的水稻机器人路径规划  

Path Planning of Rice Robot Based on Dynamic Self-Determination Parallel Fuzzy Clustering Chicken Swarm Optimization Algorithm

作　　者：余晓兰[1] 万云[1] 朱景建 陈靖照 YU Xiao-Ian;WAN Yun;ZHU Jing-jian;CHEN Jing-zhao(Chongqing City Vocational College,Chongqing Yongchuan 402160,China;Zhejiang University of Technology,Zhejiang Hangzhou 310014,China;Zhengzhou University,He'nan Zhengzhou 450000,China)

机构地区：[1]重庆城市职业学院,重庆永川402160 [2]浙江工业大学,浙江杭州310014 [3]郑州大学,河南郑州450000

出　　处：《机械设计与制造》2022年第1期251-256,共6页Machinery Design & Manufacture

基　　金：重庆市自然科学基金项目(2017CQ283)。

摘　　要：为了有效提高水稻机器人路径规划精度,提出了一种基于动态自确定并行模糊聚类鸡群优化算法的水稻机器人路径规划方法。首先,构建基于碰撞威胁度、路径长度和路径平滑度的极坐标水稻机器人路径规划模型,在降低问题求解维度的同时,提高了机器人路径规划的可行性。其次,设计动态自确定并行模糊聚类鸡群优化算法(DMCSO),该算法利用动态自确定分类个数的核FCM对鸡群进行聚类分析,并在MPI并行架构下执行协同进化操作,以提高算法求解高维复杂问题的优化性能,经典测试函数对比结果表明,DMCSO算法无论是在收敛精度上还是在运算效率上都要优于其它算法。最后,利用DMCSO算法对路径规划模型进行求解,以获得更为满意的路径规划方案。仿真结果表明,机器人路径规划方法更具可行性和合理性,路径长度降低了约(15.9~26.5)%,运算时间降低了约(20.3~44.3)%。In order to improve the precision of path planning of rice robot effectively,a path planning method based on dynamic self-determination parallel fuzzy clustering chicken swarm optimization algorithm is proposed.Firstly,a polar coordinate rice robot path planning model based on collision threat,path length and path smoothness is constructed,which not only reduces the dimension of problem solving,but also improves the feasibility of robot path planning.Secondly,an dynamic self-determination parallel fuzzy clustering chicken swarm optimization algorithm(DMCSO)is designed.The DMCSO algorithm uses the kernel FCM which automatically determines the number of chicken population classification,and performs the co evolution operation under the MPI parallel architecture to improve the optimization ability of the algorithm to solve high-dimensional complex problems.The comparison results of classical test functions show that DMCSO algorithm is superior to other algorithms in both convergence accuracy and operation efficiency.Finally,the DMCSO algorithm is used to solve the path planning model to obtain a more satisfactory path planning scheme.The simulation results show that the robot path planning method in this paper is more feasible and reasonable,and the path length is reduced by about(15.9~26.5)%,and the operation time is reduced by about(20.3~44.3)%.

关 键 词：机器人 路径规划 鸡群优化算法 模糊聚类 

分 类 号：TH16[机械工程—机械制造及自动化] TP242.6[自动化与计算机技术—检测技术与自动化装置]