The roles of camera-based rear vision systems and object-detection systems: Inferences from crash data

摘要

Advances in electronic countermeasures for lane-change crashes, including both camera-based rear vision systems and object-detection systems, have provided more options for meeting driver needs than were previously available with rearview mirrors. To some extent, human factors principles can be used to determine what countermeasures would best meet driver needs. However, it is also important to examine sets of crash data as closely as possible for the information they may provide. We review previous analyses of crash data and attempt to reconcile the implications of these analyses with each other as well as with general human factors principles. We argue that the data seem to indicate that the contribution of blind zones to lane-change crashes is substantial. Consequently, as decisions are made about electronic countermeasures for lane-change crashes, the first goal should be to support drivers' own perceptual capabilities by eliminating blind zones, as could be done with camera-based displays. More elaborate systems that could bypass driver perception with artificial object-detection capabilities might also be useful, because they could address crashes attributable to driver weaknesses in areas beyond basic sensation, such as attention or decision making. However, design of such systems should take into account any remaining blind zone problems.