This paper presents a stochastic model and a Monte Carlo computer simulation algorithm to combine all measurable non-linear kinematic and misalignment error components to predict the overall tooltip quasi-static robot spatial position accuracy for a gantry robot. These errors include joint position accuracies, joint misalignments, joint zero position offsets, axis directional straightness, squareness errors and kinematic coupling errors. All of these errors can be independently measured using a laser interferometer and/or other precision measuring instruments. The interation between robot joints and coupling between these error components are very complex making the determination of the overall robot spatial position accuracy difficult. In this paper a Monte Carlo computer simulation program for predicting overall robot spatial position accuracy based on a stochastic error model was developed. Finally, simulation results are compared with direct spatial accuracy test results using a computerized Theodolite system. This robot spatial accuracy qualification methodology has been accepted and recommended by RIA as part of American National Standard for robot accuracy evaluation.

This content is only available via PDF.
You do not currently have access to this content.