Biomedical Engineering Reference
In-Depth Information
+
CAVE
Locomotion
In some CAVE systems, users are able to navigate through a computer simulated
environment by making walking motions rather than by using a joystick or other
controls. This may be done either by tracking the motion of a user's legs while
stepping in place or while implementing some forms of redirected walking (e.g.,
[
82
]) or by having users walk on a traditional treadmill (e.g., [
69
]) or omnidirectional
treadmill [
22
,
25
,
96
] or even while making stepping or leaning motions on a Wii
balance board (e.g., [
44
]). But as we discussed above with current motion controllers
for desktop systems, this type of interface increases the
involvement
of idiothetic and
efferent systems, but does not necessarily provide
accurate
idiothetic and efferent
information. Of course vestibular information in such systems also tends to be quite
limited. The sensation of stepping in place is not the same as walking forward,
for example. As discussed below, this limitation is not unique to CAVEs, but also
applies to any VE system with limited movement range. Tethered HMDs, single-
screen projections, and motion-controlled desktop or gaming platforms (e.g., Wii,
Kinect) can all be configured to permit the user some small area in which to move,
but inherently prevent users from having true, natural mobility through the VE. Users
are aware at some level that large movements are not possible within the physically
available space, and this knowledge can interfere with even well-simulated motion.
When studying illusory motion (vection) in the laboratory, for example, informing
participants that their chair is incapable of moving can delay the onset of vection [
61
].
Alternatively, seating participants on a rotating stool that
might
rotate, raising their
feet from the floor, and providing subtle vibrations to the stool to haptically simulate
rotation can facilitate even weak vection effects such as auditory vection [
84
].
Uses for CAVEs
While CAVEs offer only limited range of motion, they are by far the most visually
immersive type of VE system currently available. These attributes make a CAVE
ideal for applications in which users do not need to move more than a few steps or in
which peripheral vision is crucial. A CAVE is ideal for any type of vehicle simulator,
in which the user remains seated in a cockpit and uses realistic controls to operate
the virtual vehicle. Although even in this case, vestibular and somatosensory inputs
would not be accurately simulated without some type of motion platform to generate
acceleration forces. ACAVEwould also be ideal to train users who will subsequently
be operating in a confined space, such as medics gathered around a patient, astronauts
working inside a space station capsule, an aircraft technician learning to repair a
virtual jet engine, or a machine operator learning to run a dangerous press on a
factory floor. Like desktop VEs, CAVEs are also well suited to visualizing small
spaces such as a prototype shelf layout in a retail store, or single virtual objects
such as a sculpture or an engine design. In these cases, extended foot navigation is
unnecessary. Indeed, it has become increasingly common for large manufacturers to
