基于入侵性杂草优化的机械手逆向运动学求解  

Inverse Kinematics Solution of Manipulators Based on Invasive Weed Optimization

作　　者：杨雨佳 神显豪[2] 刘珊珊 程日光 YANG Yujia;SHEN Xianhao;LIU Shanshan;CHENG Riguang(School of Computer Science&Engineering,Guangzhou Institute of Science And Technology,Guangzhou,510540,Guangdong,China;Guangxi Key Laboratory of Embedded Technology and Intelligent System,Guilin University of Technology,Guilin 541004,Guangxi,China;School of Computer Science&Engineering,South China University of Technology,Guangzhou,510006,Guangdong,China;School of Materials Science and Engineering,Guilin University of Electronic Technology,Guilin 541004,Guangxi,China)

机构地区：[1]广州理工学院计算机科学与工程学院,广东广州510540 [2]桂林理工大学广西嵌入式技术与智能系统重点实验室,广西桂林541004 [3]华南理工大学计算机科学与工程学院,广东广州510006 [4]桂林电子科技大学材料科学与工程学院,广西桂林541004

出　　处：《济南大学学报（自然科学版）》2021年第3期265-270,共6页Journal of University of Jinan(Science and Technology)

基　　金：国家自然科学基金项目(61741303);广东省教育厅高校青年创新人才项目(自然科学)(2016KQNCX2354);广西自然科学基金项目(2018GXNSFAA294061);广西重点研发计划项目(2017AC05027)。

摘　　要：为了有效提高机械手控制中逆向运动学求解的速度和准确度,提出一种基于极限学习机-入侵性杂草优化的机械手逆向运动学求解方法;建立工业六自由度机械手的正向运动学模型,并利用训练较快的极限学习机计算机械手关节角度向量,即输出机械手关节逆运动学初解;利用入侵性杂草优化算法对得到的初始逆解进行优化,取最小适应度时的杂草位置作为输出,以便得到最佳的逆向运动学求解。结果表明,相比于基于粒子群优化-反向传播神经网络的求解方法,所提出方法的机械手末端执行器的精度更高,平均计算时间仅为5.68 ms,实时性更好。To effectively improve the speed and accuracy of inverse kinematics solution in manipulator control,an inverse kinematics solution method of manipulators based on invasive weed optimization was proposed.Firstly,the forward kinematics model of industrial manipulators with 6-degree of freedom was established.The joint angle vector of manipulators was calculated by using the extreme learning machine algorithm with fast training speed,i.e.the initial inverse kinema-tics solution was output.Then,the invasive weed optimization algorithm was used to optimize the initial inverse solution.The weed position with the minimum fitness was taken as the output in order to obtain the best inverse kinematics solution.The results show that compared with the solution method based on particle swarm optimization-back propagation neural network,the manipulator end actuator of the proposed method has higher accuracy.The average calculation time is only 5.68 ms,and the real-time performance is better.

关 键 词：六自由度机械手 入侵性杂草优化 极限学习机 运动学模型 逆解 

分 类 号：TP241[自动化与计算机技术—检测技术与自动化装置]