Rear-end collision is the most frequent collision type, which usually occurs due to leading vehicle’s sudden brake and following driver’s late response. To reduce the collision rate, Collision Avoidance System (CAS) has been widely used to enhance road safety. The CAS used a series of algorithms to trigger alert and/or brake control to help drivers avoid collisions. However, some key variables such as the deceleration rates of leading and following vehicles were often considered as fixed values in the algorithms. Also, the detection of vehicle’s deceleration might take seconds which could extent the judgement time of CASs. The study aimed to explore drivers’ deceleration behavior pattern in car-following. A driving simulator experiment was conducted to collect drivers’ behaviors in the process. Forty-six participants took part in the experiment and the leading vehicle was designed to decelerate with different deceleration rates. The results show that change of deceleration rate was a key variable that can represent drivers’ deceleration urgency. The change of deceleration rate of the following vehicle was significantly correlated with the leading vehicle’s deceleration rate. More than 90% drivers reached a maximum change of deceleration rate within 0.5 seconds since they started to decelerate. The study implies the potential of using change of deceleration rate as an indicator in CAS to monitor vehicle’ deceleration.

CICTP