Examining the Effects of Visibility and Time Headway on the Takeover Risk during Conditionally Automated Driving

Abstract

The objective of this study is to examine the effects of visibility and time headway on the takeover performance in L3 automated driving. Both non-critical and critical driving scenarios were considered by changing the acceleration value of the leading vehicle. A driving simulator experiment with 18 driving scenarios was conducted and 30 participants complete the experiment. Based on the data obtained from the experiment, the takeover reaction time, takeover control time, and takeover responses were analyzed. The minimum Time-To-Collision (Min TTC) was used to measure the takeover risk level and a binary logit model for takeover risk levels was estimated. The results indicate that the visibility distance (VD) has no significant effects on the takeover control time, while the time headway (THW) and the acceleration of the leading vehicle (ALV) could affect the takeover control time significantly; most of the participants would push the gas pedal to accelerate the ego vehicle as the takeover response under non-critical scenarios, while braking was the dominant takeover response for participants in critical driving scenarios; decreasing the TCT and taking the appropriate takeover response would reduce the takeover risk significantly, so it is suggested that the automation system should provide the driver with the urgency of the situation ahead and the tips for takeover responses by audio prompts or the head-up display. This study is expected to facilitate the overall understanding of the effects of visibility and time headway on the takeover performance in conditionally automated driving.

Are Warnings Suitable for Presentation in Head-Up Display? A Meta-Analysis for the Effect of Head-Up Display Warning on Driving Performance

“…For the first moderator variable, we summarized the scenarios as described in the articles into several categories: following a leading vehicle event, turn left or right event, curves ahead event, lane change event, the leading vehicle brake event, obstacles ahead event, and pedestrian crossing the road event. With reference to Peng et al ( 46 ) and Wu et al ( 47 ), we then set the urgency levels of these scenarios into: critical scenarios (the leading vehicle brake event, obstacles ahead event, and pedestrian crossing the road event), non-critical scenarios (following a leading vehicle event, turn left or right event, curves ahead event, and lane change event) and mixed scenarios (containing at least one of the critical and non-critical scenarios). As there was only one articles with the mixed scenario, it was not included in the analysis.…”

“…The potential impacts of these different scenarios on the relationship between warning types and driving performance are still unknown. Peng et al ( 46 ) group the various scenarios as critical scenarios and non-critical scenarios. In non-critical scenarios, failure to intervene in time may have no direct impact on driving safety.…”

“…However, in critical scenarios, failure to intervene in time may lead to traffic accidents ( 47 ). Peng et al ( 46 ) found that the takeover control time varies considerably in different scenes, especially the shorter takeover control time needed in critical scenarios. According to Wu et al ( 47 ), the brake reaction time, steering reaction time, and takeover reaction time of the urgency scenario are significantly shorter than the semi-urgency scenario.…”