Improving Accuracy of Robotic Operations Through Detailed Robot Cell Calibration

Abstract

Robot cell calibration ensures that all areas in the robot cell, including tool pick up points and workspace tables, are calibrated with the robot. Often times, the solution to achieve higher accuracy with robot cell calibration is through manual programming which is time consuming and has limited accuracy. This thesis proposes a solution using a laser tracker to quickly calculate highly accurate workspace coordinate frames in a MATLAB algorithm to achieve a desirable robot cell calibration. The method uses 17 data points on and around the robot and calculates the coordinate frame using direction vectors and planes. Validation of this method is performed by comparing accuracy, precision and execution time to another traditional cell calibration method, the 3-point method. The results show that this method, compared to the 3-point method, excels in all three of these categories and is especially effective in environments that continually need highly accurate workspace coordinate frames.