Solving Complex Motion Planning Problems by Combining Geometric and Physical Models The Cases of a Rover and of a Dextrous Hand 1994

Abstract:  This paper addresses motion planning for robots subjected to strong physical interaction constraints. More precisely, we will show how to solve two particular instances of this general problem: the case of a rover moving on a hilly three dimensional terrain, and the case of a dextrous hand executing a grasping operation. In addition to the high dimension of the associated search spaces, such motion planning problems exhibit new difficulties coming from the fact that the dynamics of the robot/ environment interactions may strongly modify the characteristics of the generated motion. Indeed, parameters like friction, inertia, or accelerations generate physical phenomena like sliding or skidding which may obviously affect the final behaviour of the robot. Therefore, such physical characteristics have to be carefully modeled and integrated into the motion planning paradigm. The purpose of this paper is to show how this goal can be achieved by appropriately combining geometric and physical m...