A Hybrid Human–Vehicle Dynamic Model for Motion Sickness Estimation in Intelligent Driving Scenarios

With the development of intelligent driving technology, there are more and more complaints about motion sickness from passengers in autonomous vehicles. A precise tool is needed to measure ride comfort. This paper proposes a motion sickness estimation method that combines vehicle dynamics with human factors. It integrates the six-degree-of-freedom SVC model into a comprehensive human-car interaction framework. Real-world experiments collected relevant data from in-vehicle tests, the average accuracy rate of head linear acceleration prediction is 89.1%, and the accuracy rate of angular velocity prediction is 84.6%. The heart rate variability data and participant feedback are consistent with the model prediction. This consistency means the method can capture real passenger motion perception. It helps to improve the ride comfort of autonomous driving.