论文部分内容阅读
提出了一种简单实用、成本较低的基于平面约束的机器人误差补偿方法,首先利用改进的D-H法建立机器人运动学模型并通过微分变换原理得到误差传递雅可比矩阵,通过控制机器人末端执行器对标准平面进行示教,根据所测点理论上都处于同一平面这一特性建立机器人参数误差辨识模型。为避免辨识雅可比矩阵出现奇异而导致模型求解不完整,采用了Levenberg-Marquardt算法对普通的最小二乘法进行修正,对误差模型进行求解。最后将求解的参数误差补偿到控制器中从而提高机器人定位精度,并通过试验验证了该方法的有效性。
A simple and practical method of robot error compensation based on plane constraints is proposed. First, a modified DH method is used to establish the kinematic model of the robot and the Jacobian matrix is obtained through the principle of differential transformation. By controlling the robot end-effector pair, Standard plane teaching, according to the measured points are theoretically in the same plane this feature to establish a robot error parameter identification model. In order to avoid the singularity of the Jacobian matrix and lead to the incomplete solution of the model, the Levenberg-Marquardt algorithm is used to correct the ordinary least square method, and the error model is solved. Finally, the error of the parameters solved is compensated to the controller so as to improve the robot positioning accuracy. The validity of this method is verified through experiments.