Emergency Steering Evasion Assistance Control Based on Driving Behavior Analysis

Abstract

Research on collision avoidance has received significant attention in academia, but few attempts have been made regarding the impact of the real driver. Focusing on the hybrid vehicle controlled by a real driver, this paper proposes a combined yaw moment and steering torque control method based on driving behaviors during the emergency steering evasion (ESE). First, a CarSim vehicle model and a vision-based driving simulator are built to acquire and analyze ESE driving behaviors, and the trigger condition of the assistance is defined. Second, a preview distance adaptive driver steering model is created and followed by the calculation of the desired value of the steering angle and yaw rate from the planned collision avoidance path. Third, an ESE assistance controller is designed on the basis of the electric power steering (EPS) torque assistance law during normal driving and driving behaviors for the steering evasion lane change. Controllers used for this paper include a yaw rate tracking controller and a steering torque assistance fuzzy controller. The actuators are EPS, an electric stability program, and two hub motors in the rear axle. The preview distance is optimized by the particle swarm optimization method and the driving behavior-based ESE controller is tested with the driving simulator. The result shows that the optimized preview distance is credible, and the path tracking accuracy and the vehicle stability are improved with the help of the proposed ESE assistance controller by giving full consideration to driving behaviors in ESE.