A study of the effect of monitoring request and takeover request time interval on takeover characteristics

Abstract With the advancement of Level-3 conditional automated driving technology, drivers are increasingly engaging in non-driving related tasks (NDRTs) while the vehicle operates autonomously. However, this behavior can reduce takeover performance and compromise driving safety. The monitoring request (MR) mechanism alerts drivers in advance to resume monitoring traffic conditions, thereby improving takeover performance and ensuring safety. This study investigated the effect of four time intervals—3 s, 5 s, 7 s, and 9 s—on driver takeover characteristics, including takeover reaction time, maximum longitudinal deceleration, minimum collision time, fixation count, and the percentage of area of interest (AOI) fixation time. Results showed that longer time intervals significantly enhanced driver takeover performance and situational awareness. In particular, the 7 s and 9 s intervals resulted in significantly better performance and situational awareness compared to the 3 s and 5 s intervals, yet the difference between 7 s and 9 s was not statistically significant. To determine the optimal interval, the CRITIC-TOPSIS evaluation model was applied, which ranked the 7 s interval as the most effective, followed by 9 s, 5 s, and 3 s. These findings indicate that the 7 s interval achieves the best balance between driver preparedness and takeover performance, while the 3 s interval performed the worst. This study provides valuable insights into the design of the time interval between monitoring requests and takeover requests in automated driving systems, contributing to improved driving safety and user satisfaction.