Detection of Motion Sickness
Motion sickness is a problem that mainly affects passengers. A large number of influencing factors can also affect drivers individually during phases of ride-along in an automated vehicle. The goal is therefore to develop AI (Artificial Intelligence) algorithms and HMI (Human-Machine Interface) that are able to detect motion sickness on a situation-specific basis and reduce its occurrence.
Motion sickness occurs primarily in passengers while performing visual activities, such as reading, and can have an impact on a driver’s ability to take over after a period of automated driving. Consequently, motion sickness may limit the utility of automated driving for a significant portion of the population.
Therefore, predicting, detecting and counteracting motion sickness is an important task for the future. Due to the very large amount of influencing factors and large differences in susceptibility between individuals, time points and situations, the prediction and detection of motion sickness has a very high degree of complexity. Hence, the work package addresses the central question of how AI-based applications could help passengers in future vehicles to be productive and to use entertainment media without experiencing motion sickness.
To address the high degree of complexity of motion sickness, this work package will test the use of AI methods and develop an MMI that provides AI-based, situation-adaptive optimal support for the individual user.
Since driving on rural roads is likely to be a significant part of everyday motion sickness due to the curvature and unevenness of the roads, the project focuses in particular on this use case.
The goal of reducing motion sickness is to be achieved by a vehicle-integrated prototype of an AI-based visual activity manager that makes the appropriate recommendations and takes the right actions at the right time. In the work package, the proof-of-concept of an AI-based approach for predicting, detecting, and counteracting motion sickness will be designed, which, when scaled appropriately in a vehicle, can both minimize the multi-causality of motion and enrich the user experience.