Investigating the Human Impact of Perceptually Optimized Motions for Immersive Telepresence Robots

Thesis event information

Date and time of the thesis defence

Place of the thesis defence

IT115, Linnanmaa

Topic of the dissertation

Investigating the Human Impact of Perceptually Optimized Motions for Immersive Telepresence Robots

Doctoral candidate

Master of Science Katherine Mimnaugh

Faculty and unit

University of Oulu Graduate School, Faculty of Information Technology and Electrical Engineering, Center for Ubiquitous Computing

Subject of study

Human-Computer Interaction

Opponent

Professor Mark Billinghurst, University of South Australia

Custos

Professor Steven M. LaValle, University of Oulu

Visit thesis event

Add event to calendar

Investigating the Human Impact of Perceptually Optimized Motions for Immersive Telepresence Robots

Telepresence is the ability to be in a remote location using technology. Robotic telepresence through a Virtual Reality (VR) Head-Mounted Display (HMD), referred to as immersive robotic telepresence, increases the field of view for the person aboard the telepresence robot and thus improves the illusion of being present at the robot's location. Immersive robotic telepresence is used for many applications, including remote education, remote exploration, search-and-rescue, clinical care, therapeutics, and assistive technology for aging and physical limitation. However, a significant barrier to the further use of this technology is VR sickness, a constellation of symptoms similar to motion sickness which can result from VR HMD use. Other aspects related to the motions of the telepresence robot, such as the forward speed and number of turns, can also have a positive or negative impact.

To address these concerns, this dissertation investigates user experience in immersive robotic telepresence. First, a mathematical framework for robot motion planning that accounts for user comfort criteria, called "Human Perception-Optimized Planning" (H-POP), is introduced. Robot paths created with H-POP were compared to a path created with a traditional path planning algorithm, the Rapidly exploring Random Tree (RRT). VR sickness and other aspects of the paths were examined to determine how these factors impacted the users. One of the optimized paths and the RRT path were then used to induce VR sickness in another set of participants while their brain activity was measured. Self-reported levels of VR sickness were found to be associated with a measure of cognitive function, the P3b, and the amount of errors on a secondary task, indicating that VR sickness induced in current generation HMDs had quantifiable impacts on cognition and performance. Changes in the P3b can be used to evaluate many types of VR experiences, and H-POP can be used for motion planning for any type of vehicle, so both provide value beyond the scope of this work.

In sum, this dissertation provides three major contributions to the research and development community. First, it offers a framework for motion planning which accounts for user comfort criteria. Second, it discusses an evaluation of different features, such as the turns and distance from objects, that affect user experience. Finally, it provides evidence of the deleterious consequences of VR sickness on attention and performance. These results can inform developers of immersive telepresence so that VR sickness can be ameliorated and the comfort and enjoyment for users can be improved.
Last updated: 15.10.2024