Headtrack: Tracking head orientation using wireless signals

Estimating and tracking the head orientation of the user is an important problem for numerous mobile computing applications. Current solutions to the problem require deploying infrastructure (namely, cameras and lasers) along with expensive (IMU) sensors. These infrastructure-based approaches bound the user to a limited area of tracking and also disable the portability and mobility of the user. This work presents HeadTrack and explores the feasibility of designing a necklace-like wearable consisting of a headset and a chest-piece that can be used to estimate the user's head orientation using wireless (radio frequency) signals. The core problem presented in this thesis is to accurately estimate multiple distances between the chest-piece on the torso and the headset using the ultra-wide band (UWB) radios. Such a wearable not only enables portability but also mobility by decoupling the user's head motion from the body motion. Although the ultra-wide band radios have a 1GHz bandwidth and high-speed clocks, they are unable to do sub-centimeter ranging. We improve the typical ~10cm accuracy of the UWB radios by introducing a wired path between the transmitters and the receivers to serve as a reference point. We split the signal at the transmitter and route it through the wired as well as the wireless paths to improve the accuracy to about 5mm. We use ViCon to collect the ground truth for our experiments and evaluate our system. HeadTrack uses an IMU to resolve the phase wrap ambiguities and is able to track the head orientation of the user with an accuracy of 6.5 degrees. HeadTrack provides a wearable, occlusion-free, portable, and cost-effective solution to the problem of head orientation tracking with a bounded and non-diverging error.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Computer Science

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2019

Author dc:creator

• Agrawal, Amod Kant

Contributors dc:contributor

• Choudhury, Romit R.

• ultra
• wide
• band
• UWB
• tracking
• head
• orientation
• radio
• frequency
• ultrawide
• mobile computing
• RF sensing
• sensing
• augmented
• virtual
• reality
• wearable
• mobile
• headtrack
• localization
• ranging
• yaw
• pitch
• roll
• AR
• VR
• earables
• earable
• portable
• vicon
• distance
• integer
• ambiguity
• inertial
• measurement
• unit
• IMU

Statement dc:rights

• Copyright 2019 Amod Kant Agrawal

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/105242

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/105242