一种六自由度机械手动态参数辨识的研究  

Research on Dynamic Parameter Identification of a 6-DOF Manipulator

作　　者：祁晓玲 张晓华 QI Xiao-ling;ZHANG Xiao-hua(Shanxi Railway Vocational and Technical College,Shanxi Taiyuan 030013,China;School of Mechanical Engineering,Zhongbei University,Shanxi Taiyuan 030013,China)

机构地区：[1]山西铁道职业技术学院,山西太原030013 [2]中北大学机械工程学院,山西太原030013

出　　处：《机械设计与制造》2023年第12期223-229,共7页Machinery Design & Manufacture

基　　金：2021年山西省教育科学“十四五”规划项目(GH-21298);山西省高等学校重点科研项目(19B520125)。

摘　　要：针对六自由度机械手动态参数的辨识,提出了一种用于动态参数识别的激励轨迹优化方法。采用逆动力学模型和最小二乘(LS)估计方法估计机械臂的动态参数,通过改进型的傅里叶级数产生持久的激励轨迹,在保留基本动力学参数的基础上,利用Hadamard不等式对动态参数进行估计,简化了优化过程的计算时间。通过对六自由度TX-90机器人进行了模型验证,利用扭矩预测精度验证了动力学参数的可靠性,并与现有方法进行了比较,结果表明所提方法的预测均方根误差较小,优化计算时间得到大幅度减少。Aiming at the dynamic parameter identification of 6-DOF manipulator,an excitation trajectory optimization method for dynamic parameter identification is proposed.The inverse dynamic model and least square(LS)estimation method are used to estimate the dynamic parameters of the manipulator.The persistent excitation trajectory is generated by the improved Fourier series.On the basis of retaining the basic dynamic parameters,the dynamic parameters are estimated by Hadamard inequality,which simplifies the calculation time of the optimization process.Through the model verification of the six degree of freedom TX-90 robot,the reliability of the dynamic parameters is verified by using the torque prediction accuracy,and compared with the existing methods.The results show that the prediction root mean square error of the proposed method is small,and the optimization calculation time is greatly reduced.

关 键 词：动态参数辨识 轨迹优化 六自由度机械手 HADAMARD不等式 

分 类 号：TH16[机械工程—机械制造及自动化] TP2734[自动化与计算机技术—检测技术与自动化装置]