The "Darwin" Driver Vision Support System: Its Potential Impact on Driving Behaviour and Road Safety, in Conditions of Reduced Visibility
BARHAM P, Cranfield University, UK, ANDREONE L, Centre Ricerche Fiat, Italy, ZHANG X H, Vertex and VACHE M, CEDIP, France
Every driver has experienced the problems associated with driving in conditions of reduced visibility, such as night, heavy rain and fog. European car accident figures show that approximately 37% of accidents in Europe - which contribute to the 50 000 dea
Every driver has experienced the problems associated with driving in conditions of reduced visibility, such as night, heavy rain and fog. European car accident figures show that approximately 37% of accidents in Europe - which contribute to the 50 000 deaths that occur on Europe's roads each year - occur in conditions of reduced visibility. The primary effect of fog on driving is to reduce visibility distance - it would appear to make intuitive sense, therefore, that fog must present a hazard on the road. Whilst it might be hypothesised that this reduction in visibility will normally lead to a compensatory adjustment in speed in order to reduce the accident risk so presented, there is evidence to the contrary in the research literature; Sumner et al (1977), for example, found that speed adjustments made in conditions of poor visibility were, in the case of more than half of the drivers studied, insufficient to enable them to stop within their visibility distance. The authors suggest that this explains why the percentage of fatalities and seriously injured persons in the UK, as well as the number of vehicles involved in each fog-related incident, are proportionately higher in fog than in other conditions. Similarly, Snowden et al (1998) argue that, because fog reduces the contrast of the road scene ahead, so removing many visual cues as to the car's velocity, the apparent speed of the vehicle slows, which, in turn, causes the driver to actually drive faster than usual. The phenomenon of a visual scene appearing to slow down has been demonstrated by these researchers in a laboratory using a virtual environment designed to simulate the view from a moving vehicle. Test subjects were asked to "drive" at a certain speed, without the aid of a speedometer, in visibility conditions that were described as "clear", "misty" and "foggy" (where the term "foggy" described the conditions of lowest contrast). The absence of a speedometer was a fairly realistic test condition, since, in foggy conditions, drivers are often reluctant to divert their gaze from the road scene ahead to the dashboard, which further exacerbates the problem of reduced visual speed cues. Snowden et a/As results demonstrated a clear increase in driving speed as the scene became more foggy.
Furthermore, simulator-based research has shown that lane keeping performance can also be adversely affected by reduced visibility, with subjects' lateral positioning becoming increasingly inconsistent and uncertain with decreasing visibility distance (Tenkink (1988)).
Association for European Transport