Planning Motion Strategies in the Contact Space under Geometric Uncertainty Constraints

Abstract

. This work presents an approach to plan robot motions in the presence of geometric uncertainty. The contacts of the robot against the obstacles are used to reduce the position/orientation uncertainty and to guide the robot towards its goal when it is possible. Consequently, two types of motions are used to construct a motion planner: free space motions ("jumps" between the obstacles) and compliant motions (motions which maintain contacts with the obstacles). An explicit geometric model for the uncertainty is used to determine the "reachability" of the goal and the intermediate subgoals which are generated when the goal is not reachable from the current configuration. Two basic functions are combined to explore the valid configuration space: (1) a contact-based potential field function is used to generate continuous motions to move the robot to the next target configuration; (2) an exploration function is used to determine possible subgoals allowing to progress towards the ultimate goa...