华南理工大学学报(自然科学版) ›› 2016, Vol. 44 ›› Issue (5): 90-95.doi: 10.3969/j.issn.1000-565X.2016.05.014

• 机械工程 • 上一篇    下一篇

基于 WLS-ABC 算法的工业机器人参数辨识

丁力1 吴洪涛1† 姚裕2 李耀1 谢本华1 陈柏1   

  1. 1. 南京航空航天大学 机电学院,江苏 南京 210016; 2. 南京航空航天大学 航空宇航学院,江苏 南京 210016
  • 收稿日期:2015-09-02 修回日期:2015-12-02 出版日期:2016-05-25 发布日期:2016-04-12
  • 通信作者: 吴洪涛(1963-),男,教授,博士生导师,主要从事机器人动力学、并联机构研究. E-mail:mehtwu@126.com
  • 作者简介:丁力(1989-),男,博士生,主要从事机器人动力学及控制研究. E-mail:NUAADLi@163. com
  • 基金资助:
    国家自然科学基金资助项目(51375230);江苏省科技支撑计划重点项目(BE2013003-1,BE2013010-2)

Parameters Identification of Industrial Robots Based on WLS-ABC Algorithm

DING Li1 WU Hong-tao1 YAO Yu2 LI Yao1 XIE Ben-hua1 CHEN Bai1   

  1. 1.College of Mechanical and Electrical Engineering,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,Jiangsu,China; 2.College of Aerospace Engineering,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,Jiangsu,China
  • Received:2015-09-02 Revised:2015-12-02 Online:2016-05-25 Published:2016-04-12
  • Contact: 吴洪涛(1963-),男,教授,博士生导师,主要从事机器人动力学、并联机构研究. E-mail:mehtwu@126.com
  • About author:丁力(1989-),男,博士生,主要从事机器人动力学及控制研究. E-mail:NUAADLi@163. com
  • Supported by:
    Supported by the National Natural Science Foundation of China(51375230) and the Key Project of Science and Technology Support Plan of Jiangsu Province(BE2013003-1,BE2013010-2)

摘要: 针对工业机器人在不带负载时的动力学参数辨识问题,提出了一种基于加权最小二乘法与人工蜂群算法(WLS-ABC)的辨识算法. 首先计及关节摩擦特性,推导出机器人动力学模型的线性形式;接着设计五阶傅里叶级数作为激励轨迹,采集辨识实验数据;然后根据文中辨识算法,采用加权最小二乘法得到待辨识参数初始解,并以蜂群为搜索单位,通过群体之间的信息交流与优胜劣汰机制找到全局最优参数;最后对得到的模型进行验证与分析. 实验结果表明,通过文中辨识算法得到的预测力矩与测量力矩有较高的匹配度,所建立的模型能够反映机器人的动力学特性.

关键词: 工业机器人, 参数辨识, 加权最小二乘法, 人工蜂群算法

Abstract: Aiming at the kinetic parameter identification of industrial robots without loads,a novel hybrid algo- rithm,which combines weighted least square method with artificial bee colony algorithm (WLS-ABC),is pro- posed.Firstly,a linear dynamic model of the robot considering the friction characteristics of joints is deduced. Secondly,a five-order Fourier series is designed to be the exciting trajectory and experimental data are collected and identified.Then,WLS is employed to obtain the initial solution of the collected experimental data.Moreo- ver,bee colony is used as a search unit to find global optimal parameters through exchanging the information and retaining the superior individual.Finally,the established model is validated and analyzed.Experimental results show that the predicted torques well match the measured ones,and that the proposed model well reflects the kinetic characteristics of robots.

Key words: industrial robots, parameter identification, weighted least square method, artificial bee colony algorithm