Personalized road feel control strategy accounting for drivers’ strength decline

Road feel is a feature of steering resistance from the road perceived by drivers. A sound road feel design can improve the comfort and stability of drivers’ steering maneuvers. However, currently available steering systems do not adapt well to the real-time changes in drivers’ muscle strength decline, which increases their strain and reduces the stability of steering after long driving. To solve this problem, a personalized road feel control strategy that accounts for drivers’ strength decline and comprises a strength decline identification layer, and a road feel control layer is proposed. On the identification layer, the drivers’ strength decline characteristics are obtained from hardware-in-loop experimental data. The recursive least squares algorithm with adaptive forgetting factor is proposed to realize an online identification of the mechanical properties of the drivers’ arms, and their mapping relationship with the strength decline characteristics is established using the long short-term memory neural network. On the road feel control layer, considering the uncertainty of the muscle model’s parameters, a personalized road feel controller is designed, where unfixed parameters are kept within an invariant set, and the optimal road feel torque is derived from the muscle’s strength decline characteristics. Experiments show that this strategy can effectively curb drivers’ muscle strength decline and improve the smoothness of the torque after long driving.