Modified nonlinear predictive control of elastic manipulators

The paper presents an approach to end point control of elastic manipulators based on the nonlinear predictive control theory. For the derivation of the predictive control law, a vector function s is chosen as a linear combination of the trading error,its higher order derivatives, and the integral of the trading error for robustness. A control law is derived by minimizing a quadratic function of the predicted value of s and control torques. This approach avoids the instability of the zero dynamics encountered in the controller design using feedback linearization and variable structure control techniques for end-point control. Simulation results are presented for a one link flexible manipulator to show that in the closed-loop system accurate end point trajectory control and vibration damping can be accomplished in the presence of uncertainty in the system parameters