PhD Dissertation Defense by TRAN Huy Vu | Enhanced Gesture Sensing using Battery-less Wearable Motion Trackers

About The Talk

Motion sensors in wearable devices are increasingly used to enable smart, context-aware applications. By virtue of being attached to different parts of a user’s body, such as wrist, finger or ears, such devices offer unique possibilities for sensing the activities of users and their interactions with their surrounding environment. As form-factors of wearable devices continue to shrink, new energy-harvesting techniques have been studied to solve the problem of powering such wearable devices. However, the inherent high latency and high power consumption of gestural processing technologies prohibit the use of energy-harvesting wearables for many interactive applications such as AR/VR interactive movement therapy, virtual sports coaching. This thesis demonstrates that it is feasible to achieve low-latency motion sensing using battery-less wearables via a combination of: (1) intelligently beamformed WiFi transmissions, harvested by a battery-less, inertial sensor-embedded, wearable device, such that the harvested power increases sufficiently to permit quasi-intermittent sensor activation; (2) low-power, inertial sensing based, gesture recognition algorithms that support early gesture recognition and accurate hand trajectory tracking.

Speaker Biography

TRAN Huy Vu is a Ph.D. candidate in the School of Information Systems, Singapore Management University, working in the area of wearable and pervasive computing under Professor Archan MISRA. He received his bachelor degree in Computer Engineering from Ho Chi Minh University of Technology, Vietnam, in 2009. His current research focuses on developing battery-less sensing systems.