Image parameters for driving with indirect viewing systems.

Abstract

Indirect viewing systems such as tv cameras can potentially support drivers under low visibility conditions or when the driver's field of view (FOV) is restricted. In three experiments, we identified the critical image parameters of such systems on vehicle control. We used a taskbattery that measured lateral and longitudinal vehicle control in both simulated and real world driving. Important parameters are magnification factor (compared to unity, a magnification of 0.5 leads to a lower course stability and overestimation of speed and distance) and FOV (increasing the FOV from 50 degrees to 100 degrees improves performance in lateral control tasks). However, the positive effects of a doubled FOV cannot outweigh the negative effects of magnification 0.5, when both factors are confounded. Less critical is the image resolution (lowering the image resolution leads to distance overestimation and degraded longitudinal control) and image update rate (rates below 5 - 10 Hz decrease lateral control). Camera viewpoint (i.e., the location of the camera) is not critical for vehicle control. Overall, we can conclude that vehicle control with an indirect viewing system is proficient when the image parameters are adequately chosen. This supports the further development of these kind of driver support systems.