Low fatigue walking-in-place locomotion technique for mobile VR using smartphone’s inertia sensors
Mobile Virtual Reality (VR) headset that utilizes mobile smartphone for processing is a cheaper solution in experiencing VR immersion. However, locomotion in mobile VR is still a challenge because of the limitation to interact with the smartphone, as the smartphone is attached to VR headset. A co...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2020
|
| Subjects: | |
| Online Access: | http://psasir.upm.edu.my/id/eprint/90674/1/FSKTM%202020%2011%20IR.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Mobile Virtual Reality (VR) headset that utilizes mobile smartphone for
processing is a cheaper solution in experiencing VR immersion. However,
locomotion in mobile VR is still a challenge because of the limitation to interact
with the smartphone, as the smartphone is attached to VR headset. A common
solution is walking-in-place (WIP), which is a hands-free input method to control
locomotion inside the mobile VR environment. WIP uses inertia sensors in a
smartphone such as accelerometer and gyroscope to capture the inertia data
generated by the WIP gesture.
This thesis introduces Swing-In-Place (SIP) implementation that addresses three
VR locomotion research problems, which are: reducing the fatigue level of WIP
locomotion, enable viewing to different direction while moving forward, and
reducing the fatigue caused by speed controlling during WIP locomotion.
First, in order to achieve a low fatigue level WIP technique, this thesis proposes
a gesture, SIP which is less tired than the common jogging gesture used by WIP
implementation in mobile VR environment. The SIP gesture generates
acceleration by raising one foot and leaning the body to opposite site to create
horizontal impulsive force. Bilateral Horizontal Impulse (BHI) detection algorithm
is introduced to detect the positive and negative impulsive force captured from yaxis
of accelerometer. Experiment results show that there is a significant
difference between the fatigue level of SIP and jogging gesture-based WIP
implementation with a significance level of 0.001 using paired t-test, where SIP is
reported to have lower fatigue level.
Secondly, the steering direction of WIP techniques in mobile VR environment is
commonly controlled based on the user’s gaze direction because the available sensors in a smartphone are limited. However, in reality we may walk and look to
different directions at the same time. Thus, this thesis presents a walking-in place
method with the “Side View” feature, Side Viewing-enabled-Swing-In-Place
(SV-SIP), which can detect the following situations: (1) user performs SIP
gesture while looking to the front direction and (2) user performs SIP gesture
while looking to the left or right directions. A Cross-Axis Cross-Sensors (CACS)
algorithm is introduced to capture different situations using different axes input
from accelerometer and gyroscope. Experiment results show that significant
differences were found between the SV-SIP and a gaze-directed WIP
implementation for the time taken to complete the side view task and the fatigue
level using paired t-test, with a significant level of 0.02 and 0.01, respectively.
The results show that the SV-SIP implementation can increase efficiency of side
view task as compared to gaze-directed WIP implementation, and the fatigue
level of SV-SIP is lower than the gaze-directed WIP implementation.
Finally, WIP techniques for mobile VR typically use step frequency to control the
locomotion speed, which will cause the user getting fatigued easily. This thesis
proposes the Pace Switching-Swing-In-Place (PS-SIP) method to reduce the
fatigue level of speed control during WIP locomotion. The PS-SIP method uses
amplitude of body movement to switch the locomotion speed. The Amplitude
Pace Switching (APS) detection algorithm is introduced to detect the amplitude
of the user’s body movement for pace switching. The fatigue level of PS-SIP
method is reported to be lower than step frequency speed control method.
Significant difference was found between the fatigue level of the two methods
using paired t-test with a significance level of 0.002 for quantitative measurement
and significance level of 0.01 for qualitative measurement. |
|---|