Exploiting the Redundancy of a Hand-Arm, Robotic System

In this report, a method for exploiting the redundant degrees of freedom of a hand-arm mechanical system for manipulation tasks is illustrated. The basic idea is to try to take advantage of the intrinsic capabilities of the arm and hand subsystems in terms of amplitude of motions, different velocity limits and degrees of precision for the achievement of a particular task. The Jacobian transpose technique, a well-known algorithm for the solution of the kinematic inverse problem, is at the core of the proposed method for the control of the hand-arm system. Different behaviors of the hand and of the arm are then obtained by means of constraints in Null(J) added to the solution given by the Jacobian transpose method. The constraints are generated by non-orthogonal projection matrices, computed on the basis of the behavior desired from the system, without resorting to extended task space techniques. Comments about the computation of the constraints, and how to take advantage of them, are reported in the paper, as well as a description of the experimental activity currently in progress on