For the development of a Road and Weather Monitoring system, the following aspects are of major importance: a) functional design; b) risk evaluation under bad weather; and c) reliable road-weather forecasting. The first aspect deals with the overall architecture of the system. It will include road status monitoring and road condition forecasting. It will treat the combination of different weather hazards and the link with road and traffic databases. Moreover, it will deal with the relationship with the DRIVE objectives, i.e. less impact on environment, less congestion and improved traffic safety. In this respect, studies show that the road quality is not correctly evaluated by the driver, especially when the surface becomes slippery, visibility becomes poor, or gusts of wind occur. The increased risk can be determined by comparing the calculated resulting vehicle behaviour with safety limits, for example for course deviation, braking distance, etc. This risk depends on speed, density, road geometry and environment. The obtained relationships can be used to derive adjusted traffic conditions for a minimum risk. The last item, road-weather forecasting, deals with two new techniques tailored for use at road sites, one to predict aquaplaning and one to predict fog. This paper provides a brief description of the system developed within the project, to extrapolate rain intensities, measured by weather-radar, to rain intensity at dangerous spots along the road. A relation between rain intensity and the risk of aquaplaning will be shown, taking into account the road condition (slope, friction of pavement, etc). For the forecasting of low visibilities due to fog, a Knowledge Based System (KBS) has been developed, in which expert knowledge is used in combination with physical and statistical data. The system contains the meteorological heuristics, including a one dimensional physical radiation fog model. The output consists of figures of risk to exceed pre-defined visibility limits at specific sites.
Abstract