基于布谷鸟算法和牛顿法组合算法的6R机器人运动学逆解  

Inverse Kinematics Algorithm for 6R Robot Based on Cuckoo-Quasi-Newton Hybrid Algorithm

作　　者：赵瑜 ZHAO Yu(Jiangsu Vocational College of Information Technology,Wuxi 214153,Jiangsu)

机构地区：[1]江苏信息职业技术学院基础课部,江苏无锡214153

出　　处：《攀枝花学院学报》2021年第5期69-75,共7页Journal of Panzhihua University

基　　金：江苏信息职业技术学院校级课题“‘胡格模式’下重构高职数学课堂的研究与实践”(JSITJY202015)。

摘　　要：6R机器人几何结构不满足Pieper准则时,无法求得封闭形式的运动学逆解,其逆运动学通常采用智能算法和数值求解。针对智能算法收敛精度较低,数值解法对初值敏感问题,提出一种由布谷鸟算法和牛顿法组合的算法,用于求解一般6自由度关节型机器人逆运动学。组合算法通过设定阈值,采用布谷鸟算法求解获得一个低精度解,并作为牛顿法的初始解,然后基于指数衰减形式动态调整学习率,采用牛顿法快速迭代得到高精度逆解。利用MATLAB仿真平台进行数字化仿真,验证了该组合算法的可行性,结果表明:组合算法求解得到的平均误差在10-9数量级,具有较高精度,能够满足工业应用的要求。The inverse kinematic problems of 6R robotics are usually solved with the numerical method when their geometry does not meet the Pieper criterion to obtain a closed kinematic inverse solution.Aiming at the low convergence precision of the intelligent algorithm and the sensitivity of the numerical solution to the initial value,an algorithm integrating the Cuckoo Algorithm and Quasi Newton Algorithm is proposed to solve the inverse kinematics of the 6R robot.In the hybrid algorithm,a threshold was set and the Cuckoo Algorithm was adopted to obtain a low-precision solution which was set as the initial solution of the Newton Algorithm.Then,the high-order inverse solution was obtained by using Newton Algorithm with exponential decay learning rate.Simulations in the MATLAB environment was carried out and their results verified the feasibility of the hybrid algorithm.The results show that the average error obtained by the hybrid algorithm is in the order of 10-9,which has high precision and can meet the requirements of industrial applications.

关 键 词：6R机器人 逆运动学 布谷鸟算法 牛顿迭代算法 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]