China Mechanical Engineering ›› 2022, Vol. 33 ›› Issue (07): 811-817,851.DOI: 10.3969/j.issn.1004-132X.2022.07.007

Previous Articles     Next Articles

Research on Geometric Error Analysis and Parameter Identification of Modular Robots

GAO Wenbin1,2;HUANG Qi1,2;YU Xiaoliu1,2   

  1. 1.School of Mechanical Engineering,Anhui University of Technology,Maanshan,Anhui,243032
    2.Anhui Province Key Laboratory of Special Heavy Load Robot,Maanshan,Anhui,243032
  • Online:2022-04-10 Published:2022-05-04

模块化机器人几何误差分析及参数辨识研究

高文斌1,2;黄琪1,2;余晓流1,2   

  1. 1.安徽工业大学机械工程学院,马鞍山,243032
    2.特种重载机器人安徽省重点实验室,马鞍山,243032
  • 作者简介:高文斌,男,1984年生,副教授。研究方向为机器人机构学、机电一体化技术。E-mail:wenbingao@foxmail.com。
  • 基金资助:
    国家自然科学基金(51605004)

Abstract: To solve the rapid compensation problem of the errors of modular robots after reconfiguration, the geometric error sources of modular robots were analyzed and divided into module parameter errors and assembly parameter errors between modules. The joint module, link module and assembly parameters between consecutive modules were described based on product of exponentials(POE)formula and homogeneous transformation. Then, the actual kinematic model of joint-link subassembly was established. Based on high-precision balls and external measurement,an identification method for module parameters and assembly parameters between modules  was proposed to identify the kinematic parameters of the subassembly. A test of parameters identification and verification for two kinds of joint-link subassemblies with typical configuration was carried out. The results show that the average positioning errors of two subassemblies after compensation are about 1/30 than that before compensation. 

Key words: modular robot, geometric error analysis, parameter identification, error compensation

摘要: 为解决模块化机器人重构后误差的快速补偿问题,对模块化机器人几何误差来源进行分析,将其划分为模块参数误差和模块间装配参数误差。基于指数积公式和齐次变换对关节模块、连杆模块及模块间装配位姿进行数学描述,建立关节-连杆子装配体的实际运动学模型。给出一种基于精密球和外部测量的模块参数及模块间装配参数辨识方法,完成子装配体运动学参数的辨识。针对两种典型构型的关节-连杆子装配体进行参数辨识和验证试验。试验结果表明,经误差补偿后子装配体的平均定位误差约为补偿前误差的1/30。

关键词: 模块化机器人, 几何误差分析, 参数辨识, 误差补偿

CLC Number: