Comparative analysis of collision-free path-planning methods formulti-manipulator systems

• Published in: Robotica 2006-11, Vol.24 (6), p.711-726
• Main Authors: Fraile, J C, Perez-Turiel, J, Gonzalez-Sanchez, J L, Baeyens, E, Perez, R
• Format: Article
• Language: English
• ISSN: 0263-5747; 1469-8668
• DOI: 10.1017/S0263574706002888

Motion planning for manipulators with many degrees of freedom is a complex task. The research inthis area has been mostly restricted to static environments. This paper presents a comparative analysis of three reactive on-line path-planning methods for manipulators: the elastic-strip, strategy-based and potential field methods. Both the elastic-strip method [O. Brock and O. Khatib, 'Elastic strips: A frameworkfor integrated planning and execution,' Int. Symp. Exp. Robot. 245-254 (1999)] and the potential field method [O. Khatib, 'Real-time obstacle avoidance for manipulators and mobile robots,' Int. J. Robot. Res. 5(1), 90-98 (1986)] have been adapted by the authors to the problem at hand related to our multi-manipulator system (MMS) (three manipulators with five degrees of freedom each). Strategy-based method is an original contribution by the authors [M.Mediavilla, J. L. Gonzalez, J. C. Fraile and J. R. Peran, 'Reactive approach to on-line path planning for robot manipulators in dynamic environments,' Robotica 20, 375-384 (2002); M. Mediavilla, J. C. Fraile, T. Gonzalez and I. J. Galindo, 'Selection of strategies for collision-free motion in multi-manipulator systems,' J. Intell. Robot Syst 38, 85-104 (2003)]. The three methods facilitate on-line path planning for our MMS in dynamic environments with collision avoidance, where the three manipulators may move at the same time in their common workspace. We have defined some 'basic motion problems' for the MMS, and a series ofsimulations has been running that will tell us how effective each path-planning method is. The simulations have been performed and the obtained results have been analyzed by using a software program developed by the authors. The paper also presents experimental results obtained applying the path-planning methods to our MMS, that perform pick-and-place tasks sharing common working areas.