Cooperative Control Framework for Human Driver and Active Rear Steering System to Advance Active Safety

Abstract

Driving safety is quite significant for human drivers, especially in the emergencies and extreme conditions. To advance the active safety of vehicles, a cooperative control framework is proposed for human driver and the active rear steering system (ARS). Human driver model is established considering different driving characteristics to simulate different drivers. Moreover, vehicle-road model is built for ARS controller design. Then, the performance indexes of active safety, i.e., lateral stability, handling performance, rollover prevention and path tracking performance, are defined. Meanwhile, constraints for ARS controller design are proposed. Furthermore, linear-time-varying (LTV) model predictive control (MPC) is applied to ARS controller design with the consideration of multi-constraints. To evaluate the effectiveness and feasibility of the cooperative control framework, three maneuvers on the low adhesion road are conducted based on the co-simulation platform of CarSim and Simulink. Testing results indicate that the cooperative control framework can assist human driver to enhance vehicle's active safety. With the application of ARS control, the performance indexes of active safety are reduced more than 50%. In addition, it also shows that the ARS system has good adaptability and effectiveness to deal with different driving characteristics.