基于Levenberg-Marquardt算法的串联协作机器人精度标定研究  

Accuracy Calibration of Tandem Collaborative Robot Based on Levenberg-Marquardt Algorithm

作　　者：李杨 金小飞 刘国锋 吴明明[1] Li Yang;Jin Xiaofei;Liu Guofeng;Wu Mingming(School of Mechanical Engineering,Anhui Sanlian University,Hefei,Anhui 230000,China;HRG International Institute(Hefei)of Research and Innovation,Hefei,Anhui 230000,China)

机构地区：[1]安徽三联学院机械工程学院,安徽合肥230000 [2]哈工大机器人(合肥)国际创新研究院,安徽合肥230000

出　　处：《黑龙江工业学院学报（综合版）》2021年第3期112-117,共6页Journal of Heilongjiang University of Technology(Comprehensive Edition)

基　　金：安徽三联学院服务机器人应用技术协同创新中心重点项目“面向配电网作业的双臂协作机器人系统研究”(项目编号:zjqr20002);安徽省教育厅高校优秀青年人才支持计划项目(项目编号:gxyq2018125)。

摘　　要：目前研究的串联协作机器人精度标定方法标定误差较大,导致定位精度较低。为解决上述问题,基于Levenberg-Marquardt算法研究了一种新的串联协作机器人精度标定方法,参照空间坐标系转换原理和运动行为参数,将多个机器人的每个关节看作标杆,机器人通过思维规划完成需要执行的运行指令,并将指令拆分,转化为坐标的形式,通过共享模式与其他机器人行为互通,保证串联协作机器人精度标定行为的连贯性,根据Levenberg-Marquardt算法对协作机器人精度标定模型结果进行优化处理,将机器人在完成任务的基础上,需要移动的范围内固定好标定板,实现精度标定。实验结果表明,基于Levenberg-Marquardt算法的串联协作机器人精度标定方法能够有效减少标定误差,提高定位精度。The current research on the accuracy calibration method of the tandem collaborative robot has large calibration errors,resulting in low positioning accuracy.In order to solve the above problems,a new accuracy calibration method for tandem collaborative robots is proposed based on the Levenberg-Marquardt algorithm.With reference to the principle of spatial coordinate system conversion and motion behavior parameters,each joint of multiple robots is regarded as a benchmark,and the robot plans through thinking.Robots ompletes the operating instructions that need to be executed,splits the instructions into the form of coordinates and communicate with other ones through the sharing mode to ensure the consistency of the accuracy calibration behavior of the tandem collaborative robot.According to the Levenberg-Marquardt algorithm,the accuracy calibration model of the collaborative robot is optimized and the calibration board is fixed within the range that the robot needs to move after completing the task to achieve accuracy calibration.The experimental results show that the accuracy calibration method of the tandem collaborative robot based on the Levenberg-Marquardt algorithm can effectively reduce the calibration error and improve the positioning accuracy.

关 键 词：LEVENBERG-MARQUARDT算法 串联协作机器人 标定原理 机器人行为 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]