Motion planning : from digital actors to humanoid robots

Abstract

The goal of this thesis is to develop motion planning algorithms for human-like figures that take into account the geometry, kinematics and dynamics of the mechanism and its environment. We propose a three-stage strategy to solve the problem of generating motions for human-like figures that manipulate bulky objects while walking. In the process, several interesting problems and their solutions are brought into focus. These problems are, 3D collision-avoidance, two-hand object manipulation, cooperative manipulation among several characters or robots and the combination of different behaviors. Our main contribution is the modeling of the automatic generation of cooperative manipulation motions. This model considers the above difficulties, all in the context of bipedal walking mechanisms. The motion planner, connected to different controllers, is used and tested in different mechanisms, both virtual and physical.