Visually Induced Motion Sickness Estimation and Prediction in Virtual Reality using Frequency Components Analysis of Postural Sway Signal

The paper proposes a method for estimating and predicting visually induced motion sickness (VIMS) occurring in a navigation task in a 3D immersive virtual environment, by extracting features from the body postural sway signals in both the time and frequency domains. Past research showed that the change in the body postural sway may be an element for characterizing VIMS. Therefore, we conducted experiments in a 3D virtual environment where the task was simply a translational movement with different navigation speeds. By measuring the evolution of the body’s center of gravity (COG), the analysis of the sway signals in the time domain showed a dilation of the COG’s area, as well as a change in the shape of the area. Frequency Components Analysis (FCA) of the sway signal gave an efficient feature to estimate and predict the level of VIMS. The results provide promising insight to better monitor sickness in a virtual reality application.