A situation and threat assessment algorithm for a rear-end collision avoidance system

Rear-end collisions are common accident scenarios and a frequent cause of these accidents is driver distraction. This paper presents a situation assessment (SA) algorithm that estimates driver distraction by continuously assessing the steering actions of the driver. A collision avoidance (CA) system is proposed, which combines the SA with a threat assessment (TA) algorithm that estimates the effort needed to avoid a collision. It is shown that the SA algorithm proposed enables the CA system to initiate earlier brake interventions when the driver is assessed as being distracted, without significantly increasing the risk of false interventions in real traffic. The CA system has been evaluated in both collision situations on a test track and during 200 driving hours in real traffic conditions.