Path optimization for multi-robot station minimizing dresspack wear

J. Kressin, Master thesis, Chalmers University of Technology Department of Signals and Systems, supervisors D. Segerdahl and T. Hermansson, examiner B. Lennartson, May 2013.

Abstract

When dealing with off-line programming of industrial robots there are sophisticated softwares available for planning of the robot paths, one being Industrial Path Solutions (IPS) developed by the Fraunhofer Chalmers Centre. One thing that is not taken into account when finding the robot paths is potential wearing on the robots’ cable dresspacks. Since dresspacks wearing out is very expensive both in material cost and cost from downtime, there is a need for incorporating dresspack wear consideration when making the automatic path planning.

This thesis addresses the problem of finding robot paths that are less damaging for the dresspack, and the result consists of three different methods for dealing with this problem. The first method involves computationally efficient restrictions of the robot joint values in order to avoid damage to the dresspack. The second method deals with the issue of finding cable configurations that are robust to movements, since only robust configurations should be used in the final sequences. Finally, the third method involves a function that measures the cable wear as a cost, to then be minimized when doing the automatic path planning.

The three methods are tested and evaluated individually on a test case in IPS. The tests show that with the cable wear consideration, the robot takes different paths with lower values of the wearing measures than the case without cables. It is concluded that with some improvements of the methods, they can be combined into a fully implementable solution.

Keywords: cable wear minimization, robot cable simulation, path planning, joint restrictions, robust cable configurations, cable wear cost function.

Acknowledgement

First of all I would like to thank my supervisors Daniel Segerdahl, MSc, and Tomas Hermansson, MSc, for their most valuable input and support throughout this thesis work. I would also like to thank Robert Bohlin, Phd, for rewarding discussions on path planning and optimization, and Johan Carlson, Phd and Director, for his guidance and support throughout the work. I would also like to express my gratitude to my supervisor and examiner from Chalmers, professor Bengt Lennartson, for his interest in this thesis. Further on, I want to thank Mathias Sundbäck, Mikael Andersson and Johan Segeborn at the Manufacturing Engineering department at Volvo Cars in Göteborg. Your welcoming attitude helped me gain invaluable insight of the challenges with dresspack wear, as well as rewarding on-site experience of the problems. Last but not least I want to express my gratitude towards everyone else at FCC that has helped me. The including atmosphere and willingness to help has really contributed a lot to my work.

Authors and Affiliations

  • Jonas Kressin, Fraunhofer-Chalmers Centre



Photo credits: Nic McPhee