华南理工大学学报(自然科学版) ›› 2015, Vol. 43 ›› Issue (3): 49-56.doi: 10.3969/j.issn.1000-565X.2015.03.008

• 机械工程 • 上一篇    下一篇

一种工业机器人动力学参数的辨识方法

丁亚东 陈柏 吴洪涛 申浩宇   

  1. 南京航空航天大学 江苏省精密与微细制造技术重点实验室,江苏 南京 210016
  • 收稿日期:2014-10-13 修回日期:2014-12-07 出版日期:2015-03-25 发布日期:2015-02-10
  • 通信作者: 陈柏(1978-),男,教授,博士生导师,主要从事医疗机械、仿生机器人、导管机器人研究. E-mail:chenbye@126.com
  • 作者简介:丁亚东(1989-),男,博士生,主要从事机器人动力学参数辨识及控制研究. E-mail: DingCamus@126. com
  • 基金资助:

    国家“863”高技术计划项目(2013AA041004);江苏省自然科学基金资助项目(BK2012798);江苏省产学研联合创新资金———前瞻性联合研究项目(BY2012011,BK2013003-10);南京市科委产学研计划项目(201204014)

An Identification Method of Industrial Robot’s Dynamic Parameters

Ding Ya-dong Chen Bai Wu Hong-tao Shen Hao-yu   

  1. Jiangsu Key Laboratory of Precision and Micro-Manufacturing Technology,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,Jiangsu,China
  • Received:2014-10-13 Revised:2014-12-07 Online:2015-03-25 Published:2015-02-10
  • Contact: 陈柏(1978-),男,教授,博士生导师,主要从事医疗机械、仿生机器人、导管机器人研究. E-mail:chenbye@126.com
  • About author:丁亚东(1989-),男,博士生,主要从事机器人动力学参数辨识及控制研究. E-mail: DingCamus@126. com
  • Supported by:
    Supported by the National High-tech R&D Program (863 Program) (2013AA041004),the Natural Science Foundation of Jiangsu Province(BK2012798) and the Jiangsu Provincial Cooperative Innovation Foud—the Prospective Joint Re-search Project(BY2012011,BK2013003-10)

摘要: 机器人动力学参数辨识方程随着关节的增多变得更加繁琐,关节数超过 4 个后,甚至难以获得准确的参数辨识方程. 为此,文中提出一种分步辨识的方法. 该方法每次辨识的惯性参数较少,辨识方程相对简单,计算量小,辨识累计误差小,能很好地实现对超过四自由度的关节机器人动力学参数的辨识. 文中还以某型六自由度关节机器人为对象,设计了辨识实验,对分步辨识方法进行了验证. 结果表明:在保有机器人关键动力学特性的情况下,文中方法能简单而有效地获得机器人的关键动力学参数.

关键词: 工业机器人, 动力学参数, 分步辨识, 辨识算法, 轨迹优化

Abstract: As the identification equations for robots’dynamic parameters become very complicated with the increase of joints,it is difficult to obtain accurate equations of dynamic parameters when a robot is of more than four joints.In order to solve this problem,a step-by-step identification method is proposed. In this method,less inertial parame-ters should be identified in each step,which results in simpler identification equation,more efficient calcu-lation and smaller accumulative error. Therefore,the parameter estimation can be achieved quite well for the robots with more than four joints. In order to validate the effectiveness of the proposed step-by-step identification method,a robot with six degrees of freedom is taken as an object to conduct the corresponding identification experiments.The results demonstrate that the proposed method is a simple but effective way to obtaining the key dynamic parame-ters of robots without changing their key dynamic characteristics.

Key words: industrial robot, dynamic parameter, step-by-step identification, identification algorithm, trajectory optimization