|
|
|
in70mm News |
|
|
|
|
|
"in70mm.com" |
Introduction to Dimension 150 Visual Display for U.C.L.A. Driving Simulator |
This article first appeared in |
|
Written by:
S. F. Hulbert and C. K. Wojcik. Institute of Transportation and
Traffic Engineering University of California, Los Angeles. Document supplied to "..in 70mm - The 70mm Newsletter by Dr. Richard Vetter |
Issue 49 - June 1997 |
|
Introduction
Visual display is the key to successful simulation and has been the subject of many intensive investigations. In the driving simulator it is desirable to present all of the normal visual environment to the driver - an unprogrammed display on a hemispherical screen in full color, fine detail, brightness, and stereoscopic vision. Unfortunately, the present state of the art in this field is still far from this goal. The following is a brief review of the most suitable techniques. Of all the methods, motion picture techniques have the best optical characteristics but they can only produce a programmed display. A number of commercially available, wide-angle, motion picture systems produce a scene large enough to fill the entire field of view of the driver as he looks through the front windshield. The better known systems are Cinerama, Cinemiracle, CinemaScope, Circlerama, Dimension 150, Todd-AO, and Jam Handy. Both Cinerama and Cinemiracle are three-projector systems with a total field of vision of 146 deg. Circlerama uses 11 projectors and 11 screens to form a 360 deg field of vision. CinemaScope, Todd-AO and Dimension 150 are single-projector systems and their field of vision are 114, 128 and 150 to 180 deg, respectively. The Jam Handy system is also a single-projector system that uses anamorphic lenses and is used in conjunction with a spherical screen, and its field of vision is 180 deg horizontal and 90 deg vertical. In wide-angle pictures it is desirable to have the same amount of pictorial information per unit area as in a smaller angle picture and the same level of brightness, since the purpose of making a larger picture is not to have it viewed from a greater distance, but rather to enlarge the visual environment of the observer. For these reasons wide-angle, single-projector systems, use larger films and more powerful projection lamps. To improve further the brightness of the picture, special high-gain screens are employed. Expected resolving power of good quality films is 50-100 lines/mm, which for 70mm film and 140 deg field of vision can produce on the screen a picture with resolution of 2-3 minutes of arc. High-gain curved screens always involve losses in image quality due to cross-illumination. The best solution to this problem is the D-150 screen. Use of stereoscopic pictures in driving simulation is possible. However, one can expect a number of problems in application of this technique. At the present state of the art, stereoscopic pictures have a narrow field of vision, and in applying the existing stereoscopic picture techniques to wide-angle projection systems, great difficulties can be anticipated in matching the images. Since the viewer is required to wear two normally polarized lenses, a certain amount of realism would be taken away. Also the presence of the curved windshield between the viewer and the screen is quite likely to cause cross-polarization of the picture and thus greatly reduce its quality. A high-fidelity driving simulator must allow the driver to make free selection of velocity and position. It is reasonably easy to vary the speed of the projector within certain limits. The lower limit is dictated by a phenomenon called flicker. Film rate must be 16 frames per second or greater to avoid flicker. The upper limit is imposed by the tensile strength of the film. Motion picture displays have no potential for varying the speeds of other vehicles relative to that of the simulated vehicle. This is a very serious shortcoming and there are no satisfactory ways of solving this problem. Conventional motion pictures show only a scene from the view-point of the camera position. This makes it impossible for the driver, for example, to change lanes. By rotating the projector about its vertical axis it is possible to make the shift along the screen to the left or to the right. To the driver this action results in a rotation of the vehicle with respect to the road. Since the path of travel has not been changed by this action the overall impression of the driver is that he is skidding. To simulate a lane change one has to take two films synchronized, one for each lane and project them simultaneously from two adjacent projectors. While the scene from one of the projectors is displayed on the screen the scene from the other projector is masked out. Changing lanes in such a case amounts to switching the scenes from one projector to another. Unfortunately this change of scenes produces a noticeable jump from one lane into another. To diminish this type of discontinuity in visual display one have to increase the number of cameras and projectors so the lane change is taken in a number of small steps instead of one large jump. In spite of these limitations the motion picture technique has to be considered as the best method of visual display for certain types of studies in which the driver is restricted to a specific path as, for example, in a ride along a mountain road. This is so because other characteristics of motion pictures, such as resolution, brightness, size of field of vision, and color are superior to those of other techniques in terms of making the visual display realistic. |
Further in 70mm reading:
|
| Go: back
- top - back issues Updated 23-04-08 |
