Output feedback PD control of robot manipulators dispense with model base on fuzzy-basis-function-network

Aimed at trajectory tracking of robot manipulators without velocity information, the current study proposed an output feedback PD control without measuring joint velocities dispense with model. The developed controller and observer were based on a Fuzzy-Basis-Function-Network, which were employed to approximate nonlinear functions in the dynamics of controller and estimate velocity of joints. All parameters of the fuzz-basis-functions and weights could be tuned online by commix arithmetic, and no offline learning phase was needed. The observer and controller require no model knowledge of robot manipulators, including inertia matrix. This control scheme is easy to be used in project by introducing PD controller. According to the standard Lyapunov theorem extension, all signals in the closed-loop are uniformly ultimately bounded(UUB). The simulation of two-link robot manipulators results show that the presented method is effective.