CLASSIFICATION OF LEVELS OF VIRTUAL REALITY MOTION SICKNESS USING MACHINE LEARNING – ASSESSMENT AND MITIGATION

Abstract Virtual Reality applications are getting attention in fields like industry, education, healthcare, manufacturing, training etc. Engaging people in a 360-degree world with a head-mounted device may cause physical and health problems. Many people cannot use Virtual Reality devices seamlessly because of motion sickness problems. Complex graphical properties included in the scene make to feel more sickness for Virtual Reality users. Virtual Reality movements include various types of geometrical transformations such as yaw, pitch, roll, rotation, translation, scaling, shearing and transformation. Study shows that continuous involvement in a VR environment starts to show symptoms of motion sickness. The motion sickness can be measured using EEG signals. Many researchers contributed to the motion sickness detection for binary classification but there are few numbers on multi-level classification and the existing has less accuracy too. Our proposed model did multi-level classification on standard EEG data and got an accuracy of 98.82% and lime local explanation has also done in this paper.