A Motion Planning Framework for Skill Coordination and Learning

摘要: Coordinated whole-body motions are key for achieving the full potential of humanoids performing tasks in human environments and cooperation with humans. We present a multi-skill motion planning learning framework which is able to address several complex coordinations as well detection imitation constraints. The composed three main parts: first, minimal set basic skills defined order achieve stepping, balance reaching capabilities. A planner then developed coordinating solve generic mobile manipulation tasks. Finally, strategies presented improving skill execution constraints from imitation. coordinate humanoid also integrates mechanisms humanlike performances realistic environments.