【作者】 于大泳； 韩俊伟； 丛大成；

【Author】 YU Da - yong, HAN Jun - wei, CONG Da - cheng(School of Mechanics and Electronics Engineering, Harbin Institute of Technology, Harbin 150001, China)

【机构】 哈尔滨工业大学机电工程学院；

【摘要】 运动学标定问题被归结为:位姿测量值与模型预测值之间的测量残差泛函最小意义下,利用高精度的三坐标测量机来获取测量残差,以达到估计几何参数实际值的目的。在这个过程中,将并联机器人的位姿与几何参数之间的非线性关系线性化,并运用线性迭代最小二乘来估计几何参数的实际值。另外,还建立了并联机器人的逆、正运动学模型,并运用牛顿-拉普森算法解决了正运动学数值解的问题。实验结果表明,利用上述最小二乘估计的几何参数能够使并联机器人的位置和方向误差向量范数分别达到了1毫米和0.5度,这说明本文提出的标定方法能够有效地提高并联机器人的精度。更多还原

【Abstract】 This paper focuses on the accuracy enhancement of parallel robots through kinematic calibration. The kinematic calibration problem is formulated in terms of a measurement residual, which is the discrepancy between the measured pose and the predicted pose. With this formulation, one is able to i-dentify actual geometrical parameters of the parallel robot from a number of poses measured by a high precision coordinate measuring machine that allows determining the parallel robots pose with respect to the base frame. In this process, the nonlinear relation between the geometrical parameters and the pose of the parallel robot is linearized, and the actual geometrical parameters are estimated by iteratively solving the overdetermined linear least squares problem. Furthermore, the inverse kinematics and forward kinematics are derived, along with the forward kinematics solved using Newton - Raphson algorithm. Experimental results indicate that the use of actual geometrical parameters enables the parallel robots accuracy to reach 1 mm for the position error vector norm and 0. 5 degree for the orientation error vector norm, and the proposed calibration method is effective in enhancing the accuracy performance of parallel robots.更多还原