Anti-Kinetosis: A Hybrid Analytics Approach for Motion Sickness Detection in Software-Defined Vehicles

Motion sickness, or kinesis, is a major issue for passenger rides, especially in autonomous vehicles, affecting user experience and comfort. Software-defined Vehicles (SDVs) introduce new possibilities for mitigating motion sickness through real-time data modeling and AI-driven prediction. This paper presents Anti-Kinetosis, an SDV-based solution utilizing Vehicle Signal Specification (VSS) to access real-time vehicle dynamics, such as acceleration and angular velocity, for motion sickness prediction. The system integrates a mathematical model and two vision-based AI models to estimate kinetosis likelihood based on vehicle motion and passenger demographics. An application is containerized for cloud deployment, with an online demo using a digital twin to simulate real-world driving conditions. Future work will integrate the solution into a lab car for validation in real environments, ensuring reliable performance and practical feasibility. This study enhances passenger comfort in next-generation mobility and lays the foundation for personalized in-car experiences and further research on motion sickness mitigation.