Optimal nonlinear controller design for set-point control of robot manipulators

This paper presents an optimal design of a nonlinear gain PD controller for the set-point control of a rigid manipulator for handling different payloads. The nonlinear gain PD controller is selected from a class of nonlinear controllers. Sufficient conditions are given for the global asymptotic stability of the closed-loop system when using a controller from this class and the gravity compensation is not exact. To achieve optimal closed-loop system performance, convicting goals and constraints are formulated as a multiobjective optimization problem and the nonlinear gains are selected using the (/spl mu/+/spl lambda/) -ES evolutionary algorithm. The benefits afforded by the proposed controller design are demonstrated via numerical simulations performed on a two-link planar manipulator under gravity and handling different payloads.