Large Motion Control of an Experimental Mobile Manipulator with Limited Sensing

ABSTRACT Two recently developed motion control algorithms for mobile manipulators are applied to an experimental system. The manipulator motions induce significant dynamic interactions with its vehicle and suspension. A conventional fixed-base controller, which typically neglects these dynamic interactions, is shown to lead to poor performance. Both mobile manipulator control algorithms, however, which account for dynamic vehicle motions caused by manipulator motions, are shown to perform well, despite practical limitations imposed on the sensory data available for control in field environments.