Calibration of Kinematic Parameters of a Robot Using Neural Networks by a Laser Tracking System

Almost the industrial robot tasks are performed by the teaching playback method. This method has problem that the laborious and time-consuming online manual teaching is inevitable. Thus, the offline teaching based on the high positioning accuracy is desired. However, a nominal kinematic model does not consider the geometric errors and non-geometric errors. Therefore, some method of calibrating precisely the parameters is required. The kinematic parameters are calibrated by minimizing errors between the measured positions and the predicted ones by nonlinear least square method. After that, the residual errors caused by non-geometric parameters are further reduced by using neural networks.