TASK PLANNING FOR DEXTEROUS MANIPULATION USING BLIND GRASPING TELE-ASSISTANCE

An approach to operator interaction for a dextrous manipulator is desc ribed, using a tele-assistance method termed blind grasping. Given app roximate initial coordinates for. the object location, the blind grasp ing approach sequences a series of dextrous finger contact movements t o establish the object geometry and candidate stable grasp locations. The approach is implemented within the framework of the preference pla nning tele-assistant task planning architecture. Geometric and semanti c models of the world are used as triggers into pre-stored lists of ac tion sequences. Actions are triggered when their pre-conditions fail, representing a state in the would that the action can correct. Failure of the action to change the world stare hierarchically triggers sub-a ctions, until atomic actions instruct lower, levels of control. For th e first prototype AMADEUS dextrous underwater gripper, it is demonstra ted that variously shaped objects centred between the open gripper fin gers can generally be grasped using simple closure. However, off-centr e displacements as small as 1 cm require five touch samples to acquire the geometry with sufficient accuracy to choose stable grasp contact points.