Control of a robotic manipulator using neural network based predictive control

The predictive control has currently become a precious tool in various domains. It is largely studied and well known in the case of linear systems. An extension of this technique for nonlinear systems control has recently been the subject of many works. Several algorithms and particularly those using fuzzy logic and neural networks have been proposed. In this work a non linear predictive control method that uses a neural model is presented. It allows the analytical determination of the control law. The main objective of this work is the application of this method to the control of a robotic manipulator.