Adaptive driver-automation shared steering control methodology of intelligent vehicle based on individualized driver behaviors

As an effective driver-automation steering system, the haptic-shared steering system (HSS) has improved vehicle safety and reduced driver burden. However, one challenge of the HSS is efficiently reducing lane departure risk, while improving driver acceptance. This paper focuses on designing an adaptive haptic guidance system via fuzzy control considering individual driver behaviors. Firstly, the driver behaviors are obtained based on the high-fidelity driving simulator experiments of 18 drivers. Secondly, a driver model combining the optimal curvature model and neuromuscular dynamics is built, and then the parameters of driver model are identified to represent the behavioral characteristics of different drivers through the particle swarm optimization algorithm. Thirdly, the fuzzy control is proposed to adjust the driver-automation authority of the HSS based on the driver behavior. Finally, the mean absolute error and root mean square error of lateral deviation and driver input torque are presented to demonstrate the effectiveness of the proposed method. The results show that relative to the HSS with fixed authority and manual driving, the designed HSS with adaptive authority can effectively improve vehicle safety and individualized driver comfort.