Biomedical Engineering Reference
In-Depth Information
involved wheelchair transport (with and without vision), and the fifth was a stationary
(non-moving) condition with vision. The results indicated that the directional return
toward the origin was much poorer when optic flow alone specified the outbound
path. Chance et al. [
3
] set up a virtual maze, in which subjects encountered target
objects along the way. Their task was to indicate the direction to these target objects
from a terminal location in the maze. The scene of the virtual maze was provided
by a HMD. Subjects controlled their motion through the mazes using one of three
locomotion modes: Walk mode, Real Turn mode, and Visual Turn mode. The results
showed that performance in the Walk mode was significantly better than that of Visual
Turn mode. In another experiment, Bakker et al. [
1
] studied orientation performance
in VE. They tested five stimulus conditions for turning: three with and two without
visual stimuli, using one three different navigation metaphors to steer rotation. Their
results showed that most accurate rotation was found when subjects turned using
their legs.
All of these studies, in addition to more recent works such as from Ruddle and
Lessels [
23
] and Suma et al. [
25
], have shown the positive effects of physical walking
on navigation tasks.
To recreate the sensations related to physical walking in virtual environments
there are theoretically three major technical issues to overcome:
Creating a sense of walking while the true position is maintained: The device thus
requires a mechanism that cancels the displacement of the walker in the real world.
Allowing the walker to change direction: The device therefore requires a mecha-
nism that cancels omni-directional displacement.
Simulation of uneven surfaces: This is required since the terrain of the real world
includes uneven surface, such as staircases. In general, the displacement of a walker
is three-dimensional.
Pioneering work toward the development of locomotion interface devices began
in 1988 [
7
] and many different prototypes have been fabricated until the present
[
13
]. The four main existing approaches to the design of locomotion interfaces can
be categorized as follows:
Sliding shoes: The walker wears specialized shoes that generate relative motion
between the foot and the floor.
Treadmill: The walker stands on a belt conveyer that moves opposite to the direction
of walking.
Foot-pad: Two platforms are applied to the feet and move in accordance with the
motion of the feet.
Robotic tiles: Movable tiles provides a dynamic platform for walking. The tiles
move opposite to the direction of walking.
The remainder of this chapter describes successively the hardware configurations
of these four methods, and discusses related technical issues and potential usages.
