Mathematical analysis and optimisation of offline programmed robot cells in industrial applications

Abstract

This thesis is dealing about correction and improvement of modern industrial robot applications. With focus on the non-linear correction of robot linear tracks, it shows a method for minimisation of the 6-dimensional error of the reached robot points in applications. This is done by static 3D-measurement of the robots tool center point in different positions on the track. After calculation of the 6-dimensional mathematical model of the linear track using cubic splines, a correction offset for arbitrary robot skid positions on the linear track can be calculated. Due to the compensation of the gap between simulated robot cell and real application, the usage of offline robot program points without the need for online correction is significantly increased. This effects time savings during setup of the robot application cells and reduces the effort in calibration and high-accurate assembling of linear track on-site.