Self-learning Adaptive Integrated Control of an Electric Vehicle in Emergency Braking

It is challenging to achieve high braking efficiency as well as high directional stability in emergency μ –split braking manoeuvres. A self-learning adaptive integrated control scheme is presented for an electric vehicle (EV) which has a novel brake-circuit configuration. A self-learning time varying super twisting sliding mode-based anti-lock braking system (ABS) controller is integrated with a simple PID-based steering controller, adaptive super twisting sliding mode-based yaw moment controller and a yaw moment allocation module via a two-tier two-layer hierarchical scheme. The ABS controller is designed based on a model which includes the actuator dynamics, and a fuzzy module is employed to vary the slope of the sliding surface to achieve high performance levels in μ –split operation. The scheme effectively executes differential braking to attain high braking performance with optimal steering effort and improved vehicle stability. Moreover, the scheme exhibits high robustness and adaptability to uncertainties and disturbances. The design has the added benefit that it is straightforward to implement in real-time. The performance of the proposed scheme is demonstrated using a 15th – order high fidelity vehicle model whose performance has been correlated with an experimental vehicle.