Search Results

Vehicle dynamics are dominated by tire forces and their precursor input variables, and a few inertial and suspension properties. Tire forces have been studied since 1930's with the purpose of comprehending and improving vehicle control and handling. Although work has been done on modeling the tire forces using various forms of polynomial interpolation of experimental data, these properties are not thoroughly recognized. As vehicle dynamics' technology improves, the more accurate tire model is required. This paper presents a tire force model by the neural networks. The neural network tire model relates the tire force as a function of vertical load, slip angle, camber angle, and longitudinal force. Many published tire model papers described the side force as Fy(α,γ,Fz)=F(α,Fz)+F(γ,Fz). However, the neural network tire model described the side force as Fy=F(α,γ,Fz) which is closer to the true tire force behavior than the conventional tire model.

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.