Biomedical Engineering Reference
In-Depth Information
10.3.3 Virtual Traveling
Instead of just implementing natural walking, many immersive VEs make use of a
hybrid walking-and-flying metaphor, in which the user's head is tracked in a limited
interaction space, whereas the user can change the reference position or orientation
in the virtual environment by using a hand-held controller. This interaction technique
can easily be grasped by users when introduced as a
flying carpet
[
40
], i.e., the user
can naturally walk over a limited carpet region, while the carpet itself can be flown
through the virtual world. In contrast to real walking, which is identified by natural
locomotion in the physical workspace, flying and steering techniques are denoted as
virtual
traveling
[
3
]. A traditional implementation of virtual traveling is view-directed
flying [
3
,
18
,
24
], which refers to user-initiated changes of reference coordinates rel-
ative to the user's virtual view, i.e., the coordinates of the virtual camera object. With
view-directed flying in immersive virtual environments usually only the reference
position is changed, whereas the orientation of reference coordinates is not affected,
such that the virtual interaction volume remains level to the real world [
18
].
A basic virtual flying controller can be implemented using the following simple
approach. For each rendering frame
n
∈ N
we compute the current view-direction
v
(
n
)
v
(
n
)
v
(
n
)
s
(
n
)
s
(
n
)
s
(
n
)
3
, the strafe-direction
3
, and the up-direction
(
,
,
)
∈ R
(
,
,
)
∈ R
v
x
v
y
v
z
v
x
v
y
v
z
u
(
n
)
u
(
n
)
u
(
n
)
3
of the camera object in the VE (see Sect.
10.2
). Providing
commodity input hardware to the user, such as a keyboard, the user can initiate
changes in reference coordinates by pressing different keys. For instance, if we
detect that the user has pressed the up- or down-key on a keyboard, we compute the
reference position for the next rendering frame as
(
,
,
)
∈ R
v
x
v
y
v
z
r
(
n
)
r
(
n
)
r
(
n
)
r
(
n
−
1
)
p
x
r
(
n
−
1
)
p
y
r
(
n
−
1
)
p
z
v
(
n
)
v
x
v
(
n
)
v
y
v
(
n
)
v
z
g
(
n
)
v
(
p
x
,
p
y
,
p
z
)
=
(
,
,
)
+
(
,
,
)
·
,
with
g
(
n
)
∈ R
defining a speed factor for virtual traveling in view direction, e.g.,
v
with
g
(
n
)
0 for forward motions if the user pressed the up-key, and
g
(
n
)
0for
backward motions if the user pressed the down-key on the keyboard. In particular,
this means that the user can turn the head towards a target in the virtual scene, and
fly towards the virtual target by pressing the up-key. Using corresponding keys and
speed factors, we can allow the user to change the reference position in the virtual
scene not only in view-direction, but also in strafe-direction
>
<
v
v
s
(
n
)
s
(
n
)
s
(
n
)
(
,
,
)
, and
v
x
v
y
v
z
u
(
n
)
u
(
n
)
u
(
n
)
up-direction
. The speed factors may be as simple as a constant or a
more sophisticated function based on sensor inputs.
(
,
,
)
v
x
v
y
v
z
10.4 Nonisometric Virtual Walking
Isometric mappings enable users to explore a virtual region by real walking. How-
ever, since with isometric mappings the virtual interaction space is limited, virtual
traveling techniques have to be used to cover larger distances in the VE, which impair
