• CN:11-2187/TH
  • ISSN:0577-6686

›› 2009, Vol. 45 ›› Issue (3): 296-301.

• 论文 • 上一篇    下一篇

扫码分享

机器人关节面时变物理参数在线识别的谐波传播法

静大海;刘晓平   

  1. 河海大学电气工程学院;北京邮电大学自动化学院
  • 发布日期:2009-03-15

On-line Identification of Time-varying Physical Parameters of Robot Joint Based on Harmonic Propagation

JING Dahai;LIU Xiaoping   

  1. College of Electrical Engineering, Hohai University Automation School, Beijing University of Posts and Telecommunications
  • Published:2009-03-15

摘要: 机器人在运动状态下关节面处的物理参数不断变化,导致机器人整体结构的动态特性也不断发生变化。为了研究机器人在运动状态下的动态特性以达到对机器人振动进行实时控制的目的,必须对机器人关节处的物理参数进行在线跟踪辨识。基于此,在现代谐波识别技术的基础上提出时变系统谐波频率与振型的在线识别算法,并利用波传播方法计算出3自由度机器人关节处的力与位移矢量,再由机器人关节面位移连续与力平衡条件求得机器人关节处的动刚度,进而求得机器人关节面处的刚度与阻尼。识别方法可以简化测试节点,可以节省测试成本。试验表明此方法具有良好的跟踪能力与较高的计算精度。

关键词: 机器人, 连接参数, 时变系统, 系统辨识, 谐波传播

Abstract: The joint physical parameters of robot change when the robot moves, which causes the change of dynamic characteristics of robot. It is very important to track and identify the joint physical parameters of robot in real time for controlling vibration of robot. An on-line identification method of frequencies and mode shapes of time-varying structure system is presented based on modern harmonic identification, and the force and displacement of joint of the robot are computed by harmonic propagation method, and dynamic stiffness and dynamic damping are computed based on displacement accord and force balance, finally the stiffness and the damping of joint of robot are presented. The cost of testing is saved because the method can predigest testing node. Experiment shows that the method has good tracking capacity and high computing precision.

Key words: Harmonic propagation, Joint parameter, Robot, System identification, Time-varying system

中图分类号: