preceding vehicle. That is, displaying the own momentary braking distance with little or no reference to the position of the preceding vehicle may simply not assist in staying out of a critical zone. This explanation is supported by the finding that the HUD-like CAS showed no increase in short headways when the preceding vehicle was actually visible, that is, in the normal visibility condition.
That the CAS with the extra 1 sec headway criterion performed worse than its counterpart--without the extra--may reflect basic complexities that arise from combining an absolute criterion with a relative one. A time-to-collision criterion takes relative speed into account, whereas a headway criterion depends only on the following vehicle's speed. Combining the two may make it hard to understand for a driver when the CAS is going to act. What drivers notice is that the distance to the leading vehicle at which the accelerator pedal starts to produce an extra counterforce varies in a complex way as a function of their own speed as well as of the relative speed with which they approach the leading vehicle. The net result of this may be that the driver chooses to disregard what the CAS is suggesting, relying instead on their own judgment.
The absence of differential CAS effects in different visibility conditions shows that a CAS may help irrespective of what perceptual problem (detecting a vehicle versus recognizing its movement) the driver has. It is, in fact, a fundamental characteristic of a time-to-collision criterion that it captures both the detection and the recognition aspect. That is, if a time- to-collision criterion is met this indicates at the same time that there is a preceding vehicle and that its relative speed is such that it will be reached within, say, 4 sec. Other criteria, including the simple headway criterion, do not have this characteristic. It may be for this reason that a time-to-collision criterion is capable of providing support to drivers even under conditions of adverse visibility.
Janssen W. H. ( 1989). The impact of collision avoidance systems on driver behavior and traffic safety. DRIVE-I. Deliverable GIDS/MAN 1.
Janssen W. H., & Nilsson L. ( 1990). "An experimental evaluation of in-vehicle collision avoidance systems". DRIVE-I: Deliverable GIDS/MAN 2.
Nilsson L., Alm H., & Janssen W. H. ( 1991). "Collision avoidance systems--Effects of different levels of task allocation on driver behavior". DRIVE-I. Deliverable GIDS/MAN 3.
Zhang X. ( 1991). Intelligent driving--PROMETHEUS approaches to longitudinal traffic flow control. Vehicle Navigation & Information Systems, Conference Proceedings SAE, 999-1010.