An On-Line Learning Control of Unsupervised Neural Network for a Vehicle Four Wheel Steering System

The dynamics of a four wheel steering(4WS) system inherently has model uncertainties, resulting in degradation in performance. As a way to compensate the model uncertainties of the vehicle system, a nonlinear neural network control scheme is proposed and evaluated. The control scheme is composed of a conventional model reference control term and a compensator term. The compensator term is generated by an unsupervised neural network whose teaching signal is just error information between the actual plant and the reference model. This control scheme does not require an inverse dynamics of the plant or a Jacobian information of the learned plant, so that an on-line learning can be carried out. Since the teaching signal of this scheme is simple to compute in the control process, the fast convergence can be realized. The validity and effectiveness of the proposed control scheme for a vehicle four wheel steering are verified by computer simulations.