抄録
A PID local feedback Scheme for the position control of robot manipulators is investigated. In addition to the ordinary use of joint-angle position and velocity feedbacks at each joint actuator, the feedback of integration of positional reference errors is introduced so as to reduce the steady state offset. In spite of the existence of strong couplings between joints, nonlinear terms of Coriolis and centrifugal forces, and bias terms due to frictions or the change of payload, the stability and robustness of the proposed scheme is proved by the Liapunov method of stability analysis incorporating a singular perturbation technique. The stability of a kind of software servoing control is also discussed. The PID feedback scheme is then modified to be suited to the case when the target point, into which the end effector of the manipulator is maneuvered, is described in terms of the Cartesian coordinates or any other task oriented coordinates. The feedback signal for this case is calculated based upon the data of sensory signals taken from both inside and outside of the manipukator, and hence the localization of feedback becomes unavailable. Nevertheless, the asymptotic stability of such a PID sensory feedback scheme is also proved under some condition on the Jacobian matrix.
