Design and Optimization of a Virtual Reality Locomotion System using Pressure Sensors

Virtual Reality (VR) is an immersive technology that enables users to interact with three-dimensional digital environments through headsets and controllers. Among its primary goals are to enhance immersion and presence, but several challenges can compromise the experience or even render it unusable. Key issues include motion sickness, device discomfort, immersive locomotion, user safety and affordability. In particular, as VR becomes more accessible for home use, locomotion presents a significant challenge. Users often lose awareness of their real world surroundings, making real walking in VR unsafe. To address this, various locomotion techniques have been developed, including joystick-based movement, teleportation, arm-swinging techniques, and seated locomotion techniques. Each one of these techniques has its own strengths and weaknesses: Joystick-based movement is intuitive but can induce motion sickness, while teleportation reduces discomfort but disrupts immersion. Arm-swinging techniques offer a more natural experience, simulating walking without actually moving. Seated techniques, exemplified by the Walking-Seat, offer an immersive, intuitive, and effortless experience thanks to the seated position. Specifically, Walking-Seat is a seated leaning-based locomotion device which allows the user to move their avatar in the virtual environment (VE) by leaning their torso. The mapping of the movement is due to the position of user's center of mass, which is calculated using four load cells. This system works has demonstrated efficacy in numerous scenarios, but presents some issues, particularly when the user needs to interact with objects placed on the ground within the VE: when the user leans forward to reach the object, the device incorrectly interprets the movement as an intention to move forward, impeding interaction and causing user frustration. In order to address this issue, the present thesis proposes two approaches: the first is to redesign the interface using high resolution devices to re-implement the Walking Seat and then optimize it to make it affordable, the second approach is to find a correlation between motion and intention. For the first approach, we opted to use a pressure mat --- comprised of a matrix of piezoelectric sensors --- and compared it with a force platform. Collecting data with a high resolution device like a pressure mat enables us to study and optimize the size and number of sensors to substitute it with. For the second approach, the analysis of the data on the correlation between movement and intention showed us some limitations caused by the way the data are acquired. For each approach, two different immersive scenarios were used to collect users' behavioral data.