Discrete-time model reference adaptive controller design for robotic manipulators

Abstract

With the trend of digital control in robotic application, two discrete-time model reference adaptive control schemes which can be implemented in an on-line microprocessor to eliminate the nonlinearities generated by the interactions of the three-link robotic manipulator are presented here. Without knowing the parameters of the system, we can achieve accurate trajectory tracking in a short time. The stability analysis stemmed from the Lyapunov's Direct Method. It guarantees the asymptotic stability of a system. The same as any discrete-time control systems, the time required for the computer to perform the necessary computation can result in extended sampling time, which may degrade the stability of the system. We obtain the sufficient condition of stability for discrete-time system. A set of right sampling time and weighting factors are chosen to show the highly accurate tracking. Through the MATLAB software simulation, we can see that the manipulator does follow the desired trajectory with high accuracy even tested by sinusoidal trajectory or with payload change. It proves the effectiveness of the designed controllers.