ROBUST ADAPTIVE CONTROL OF ROBOT MANIPULATORS

A robust adaptive control scheme is presented in this paper for the control of manipulators which have uncertainties in parameters and disturbances. The robustness is achieved by applying the Model Reference Adaptive Approach in order to obtain the adaptive laws which do not require the exact knowledge of parameters or payloads and disturbances. First the problem of independent control of individual axes is analysed. Using the second method of Lyapunov a control scheme is proposed which results in an asymptotically stable system performance. The system performance quality is provided by a suitable choice of models. When controlling of robots with n degree of freedom (n-DOF) the questions arise: • May the proposed control principle be used for every controlled axis independently; •• Or should the robust adaptive control problem be solved for the nonlinearly interconnected large scale system? The control according to the first approach is easy to realize, while the second approach guarantees the system stability. The authors solved the second problem and proposed a control algorithm. When the robot is controlled by a digital device an appropriate discrete robust adaptive algorithm can be established directly from the impulse transfer function of robot axis. Key word index. Adaptive control, Robots, Lyapunov methods, Discrete time systems.