Driver Workload for Rear-Vision Systems With Single Versus Multiple Display Locations 2005-01-0445
Advances in camera and display technology have increased interest in using camera-based systems for all rear-vision functions. The flexibility of camera-based systems is unprecedented, and raises the possibility of providing drivers with fields of view that are very different from, and potentially much better than, those of conventional rearview mirrors. Current fields of view are based on a combination of driver needs and the practical constraints of mirror systems. In order to make the best use of the greater flexibility offered by cameras, a reassessment of drivers' needs for rear vision is needed. A full reassessment will require consideration of many factors. This paper offers a preliminary analysis of one of those factors: the visual workload involved in using rear-vision systems with single versus multiple displays. The passenger car rearview mirror system that is currently typical, with three display locations, is used as a reference point against which to compare the potential effectiveness of a system with a single, unified display. Three types of data from previous studies are used: (1) an estimate of ideal driver eye fixation patterns for rear vision, (2) actual fields of view with the three mirrors that are currently typical on passenger cars, and (3) estimates of the visual angle within which drivers can take in useful information in a single visual fixation. The results suggest a considerable advantage of using a single, unified display for rear vision. All of the data used here can be improved upon, and the quantitative results described here should be considered preliminary. We present them primarily to illustrate the line of argument we are proposing, and to suggest, tentatively, that the preferred form of video rear-vision systems may in fact be markedly different from current mirror systems.