Estimating entire geometric parameter errors of manipulator arm using laser module and stationary camera

Abstract

This paper proposes a novel kinematic calibration technique to estimate entire geometric parameter errors of robot manipulator. There always exist small errors in link length and link twist for physical manipulators. These errors eventually affect the precision in kinematic equations, and cause inverse kinematic equations to calculate joint angle values having errors. In order to solve these problems, the proposed kinematic calibration technique employs a laser module and a stationary camera. The laser module made with three laser sources is attached to the end-effector of manipulator arm and the stationary camera is used to determine an accurate position of three laser beams on a 2D screen. A non-singular Jacobian matrix is formulated by using differential kinematics and transformation mapping equations to represent the variances between actual and measured positions of laser beams. Then two parameter estimation algorithms, least squares estimation and extended Kalman filter, are used to estimate the entire geometric parameter errors. Effectiveness of the proposed calibration technique is verified with 7 degrees of freedom MyBot humanoid arm by computer simulation.