Modeling of Driver's Driving Posture

This study proposes a driver posture model that incorporates direction-dependent posture control parameters to enhance prediction accuracy during vehicle operation. Real-world driving experiments were conducted on a circuit test course including alternating left and right turns by using motion capture and foot contact force sensors. The results revealed that there is asymmetry in head tilt angles, with greater inward inclination during left turns. Analyses of footrest forces indicated that this asymmetry arises from active use of the footrest pressing force during right turns to maintain posture against inertial force. To model this posture maintenance strategy, posture control of model was developed based on otolith and somatosensory feedback by incorporating gain switching according to turn direction. Model validation showed that by introducing direction-dependent gain parameters derived from footrest pressing force analysis, the model achieved improved prediction accuracy in both turning directions. These findings suggest that drivers adapt their posture control strategy in response to asymmetric foot support conditions. The proposed model contributes to realistic simulation of driver behavior and supports model-based design of human-centered vehicle systems. Future work will focus on further enhancing the model’s predictive accuracy.