VR Navigation
|
|
|
"Navigational Control Effect on Representing Virtual Environments", HFES 2003
Motivational Question
Today's computer systems make it possible to have a representation of a very large, 3D space, and to allow the user to explore it to become familiar with
it, to find regions of interest, or to make decisions that will be carried out in the real world. However, the limitations of display technology often put
constraints on the observer's visual exploration of this space. In particular, it is common to have a fixed viewport (for example, computer
monitor) in which only a small part of the total area can be seen at any one time. Navigation is then carried out as if one were driving a vehicle with a
joystick, viewing the world through a small windshield. However, without the physical movement of the vehicle, and without additional side windows, it is
very easy for the user to become disoriented, losing track of direction. This may be particularly likely if the joystick is of the 'relative effect' type,
in which moving the joystick produces motion relative to the current position.
One way of providing an absolute reference concerning direction in this situation is to employ the joystick in an 'absolute mode,' in which the
position of the joystick directly indicates the direction of the viewport relative to the world. Another option, if the viewpoint were always going to
be on the surface, would be to use a controlling object that can be moved in 2 dimensions or rotated on a position sensitive board, thus allowing it to
directly indication location and direction of the virtual vehicle and its
viewport. Both of these methods provide kinesthetic, and perhaps visual, information regarding the part of the world being seen at any given moment.
An alternative approach to navigation through the represented space is to use a helmet-mounted display with head tracking. Within this environment, the
display can be changed in response to head motion, displaying the part of the scene that would actually be in the direction to which the person is looking.
In this way, the person has normal kinesthetic feedback to anchor in space the information being seen at any moment.
The purpose of the study to be conducted is to determine the relative effectiveness of these three methods, plus viewing the world without control,
in establishing the spatial position of objects in the observer's mental representation of the space being examined.
Experimental Design
In order to control for visual characteristics of the display, a head-mounted display will be used for all conditions. Also, the observer will
remain at the same point in the space but be able to look in any direction from that point. In the relative joystick condition, movement will occur in
response to the relative motion of the joystick (pressing the joystick to the left will induce further leftward movement). In the absolute joystick
condition, the absolute position of the joystick will indicate the direction of the
viewport, essentially pivoting the viewport around to the direction
indicated. In the head-controlled condition the direction of the head will determine the direction of the
viewport. And in the control condition, the
observer will have no control over the direction of the viewport, but will see the scene in the same sequence as a subject in one of the other conditions.
Observers will be permitted to visually explore a space for 45 seconds, trying to remember the locations of objects. They will then have the display
removed, will be shown the location of a prominent object lying directly in front of them, and will be asked to point to other objects. Degree and nature
of their error will be used as data to investigate the quality and characteristics of the mental representation of the viewed space that they
have developed.
Results of this study should provide information about the relative effectiveness of these different control and display methods in examining a
space larger than can be presented on the display at once.
|
|
