CALIBRATION OF A 6-DOF SPACE ROBOT USING GENETIC ALGORITHM  

作　　者：LIU Yu JIANG Yanshu LIANG Bin XU Wenfu 

机构地区：[1]College of Science and Technology, Harbin Institute of Technology, Harbin 150001, China [2]Department of Automation, Harbin University of Science and Technology, Harbin 150008, China [3]Colloge of Science and Technology, Harbin Institute of Technology, Harbin 150001, China

出　　处：《Chinese Journal of Mechanical Engineering》2008年第6期6-13,共8页中国机械工程学报（英文版）

基　　金：supported by National Natural Science Foundation of China(No.60775049).

摘　　要：The kinematic error model of a 6-DOF space robot is deduced, and the cost function of kinematic parameter identification is built. With the aid of the genetic algorithm （GA） that has the powerful global adaptive probabilistic search ability, 24 parameters of the robot are identified through simulation, which makes the pose （position and orientation） accuracy of the robot a great improvement. In the process of the calibration, stochastic measurement noises are considered. Lastly, generalization of the identified kinematic parameters in the whole workspace of the robot is discussed. The simulation results show that calibrating the robot with GA is very stable and not sensitive to measurement noise. Moreover, even if the robot＇s kinematic parameters are relative, GA still has strong search ability to find the optimum solution.The kinematic error model of a 6-DOF space robot is deduced, and the cost function of kinematic parameter identification is built. With the aid of the genetic algorithm （GA） that has the powerful global adaptive probabilistic search ability, 24 parameters of the robot are identified through simulation, which makes the pose （position and orientation） accuracy of the robot a great improvement. In the process of the calibration, stochastic measurement noises are considered. Lastly, generalization of the identified kinematic parameters in the whole workspace of the robot is discussed. The simulation results show that calibrating the robot with GA is very stable and not sensitive to measurement noise. Moreover, even if the robot＇s kinematic parameters are relative, GA still has strong search ability to find the optimum solution.

关 键 词：Robot calibration Position and orientation accuracy Measurement noises Genetic algorithm 

分 类 号：TP24[自动化与计算机技术—检测技术与自动化装置]