Electric Power Assist Steering (EPAS)
Electrical Power Steering Systems (EPAS) are currently replacing their traditional hydraulic steering systems in vehicles. Electric Steering systems are nonlinear MIMO systems with multiple objectives, including fast response to the driver torque command, good driver feel, and attenuation of load disturbance and sensor noises. Optimal control method is employed to design such controllers for improved performance and robustness. Although those controllers have showed acceptable performance for certain operating conditions, they show some undesired steering feel for high steering gain. In this work, neural networks are set to replace the optimal controllers to guarantee robustness. A Euclidean Adaptive Resonance Theory (EART) networks is trained according to the data collected from an H∞ optimal controller. The collected data represent the controller input and output signals. The said data are normalized and clustered into categories in the EART modules. The modules are interconnected by a map field. Once the training is accomplished, the EART controller becomes ready to completely replace the optimal controller. In the classification phase, the EART controller, performs in the same manner as the original controller whose data was used for training.
Ref: This work is in collaboration with one of CARET's Advisor(s).