Contribution sur l’étude de répétabilité en position des robots industriels

Abstract

The repeatability error quantifies the level of spatial dispersion of a robot’s programmed pose. This is a very important meteorological characteristic of an industrial robot. The recommendations for calculating the repeatability of industrial robots are presented with the ISO 9283 standard. Two robots, LR Mate 200iDfrom FANUC and KR6 from KUKA, are used in this project. Their repeatability is measured by a Trical device according to the ISO 9283 method. This portable 3D measuring device can be used for calibrating the robot, evaluating its position accuracy, or simply for measuring its position repeatability. Indeed, note that the repeatability of industrial robots is typically measured with a laser tracker, but a laser tracker in not accurate enough to measure position errors in the range of 0.020 mm or less. In contrast, our Trical is accurate to a few micrometers when it comes to measuring position errors of up to 1 mm. This project presents the results and statistical analyses of a study devoted to the estimation of repeatability. The influence of three factors (warm-up, loading and flexibility) is studied. We measure repeatability infive positions according to the ISO 9283 cube, by taking 300 measurements (per position) for each different test. A two-factor, three-level Complete Factorial Design of Experiments (DOE) was adopted in these experiments. The two main factors (loading and warming up) are investigated. Three positions are chosen for further analyses, one when the robot is folded (therefore more rigid), one when it is stretched (therefore more flexible), and one in between to see the effect of flexibility on repeatability. The results show that the two robots behave globally in the same way, and the worst dispersions are presented when the robots are cold. The dispersion reaches a value of up to 0.27 mm and 0.15 mm for the FANUC and KUKA robots, respectively. After at least 2 hours of warmup, the differences are improved to reach values of less than 0.005 mm. It was also found that dispersion is strongly affected by loading during the cold period. However, this effect gradually disappears over time, especially when the robot reaches the period of thermal stability. In addition, the cold state period and the P2 and P3 positions provide the worst repeatability for the two robots (flexibility factor). After using the ANOVA analysis to see how the flexibility factor influences the two robots, we found that this factor is affecting the FANUC robot more.

Date	20 Apr 2021
Original language	French
Awarding Institution	École de technologie supérieure
Supervisor	Antoine Tahan (Supervisor) & Ilian Bonev (Co-supervisor)