融合坐标系误差模型的工业机器人运动学标定  

Kinematic Calibration of Industrial Robot Based on Error Model of Coordinate System

作　　者：陈琳[1] 乜聘广 蒙成琦 潘海鸿[1] CHEN Lin;NIE Pinguang;MENG Chengqi;PAN Haihong(School of Mechanical Engineering,Guangxi University,Nanning 530004,China)

机构地区：[1]广西大学机械工程学院,南宁530004

出　　处：《组合机床与自动化加工技术》2023年第12期178-182,共5页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：国家自然科学基金资助项目(51465005);广西创新驱动发展专项项目(桂科AA18118002)。

摘　　要：为解决辨识过程中MD-H误差模型不能完整描述机器人结构参数影响机器人位置准确度的问题,提出一种融合多坐标系的误差模型实现工业机器人运动学标定。该模型首先使用D-H模型建立机器人坐标系模型;然后,引入6参数模型描述坐标系间的齐次变换矩阵,建立融合多坐标系的误差模型;最后,使用最小二乘法进行参数辨识,将辨识得到的参数补偿到机器人结构参数中。对埃夫特ER6-C10机器人进行标定与补偿实验,并与MD-H误差模型辨识补偿后的效果进行对比。实验结果表明所提出的融合多坐标系的误差模型相比于MD-H模型将机器人平均位置误差降低了48.61%。In order to solve the problem that the MD-H error model can not fully describe the robot structure parameters affecting the accuracy of the robot end position,a multi-coordinate system error model was proposed to realize the kinematic calibration of industrial robot.Firstly,the D-H model is used to establish the robot coordinate system model.Then,a 6-parameter model is introduced to describe the homogeneous transformation matrix between coordinate systems,and a multi-coordinate error model is established.Finally,the least square method is used for parameter identification,and the identified parameters are compensated into the robot structural parameters.The calibration and compensation experiments of Eft ER6-C10 robot were carried out,and the results were compared with those of MD-H error model.Experimental results show that compared with MD-H model,the proposed multi-coordinate error model reduces the robot′s average position error by 48.61%.

关 键 词：工业机器人 参数辨识 6参数法 工具坐标系 误差模型 

分 类 号：TH165[机械工程—机械制造及自动化] TG659[金属学及工艺—金属切削加工及机床]