Risk estimator for control in intelligent transportation system

This paper introduces a two-level risk estimation system suitable to control in ITS. The top-level risk estimation is done on the basis of perceived risk associated with various driving situations and affected by weather, traffic and road conditions. The high-level risk estimation is then refined on the basis of real-time information about the vehicle surrounding, such as motions of other vehicles. The approach is illustrated on examples of maneuvers in which the risk is estimated via logic, lookup tables and neural networks.