Citation

Li M, Feng Z, Zhang W, Wang L, Wei L, Wang C. Transp. Res. C Emerg. Technol. 2023; 156: e104324.

Copyright

(Copyright © 2023, Elsevier Publishing)

DOI

10.1016/j.trc.2023.104324

PMID

unavailable

Abstract

During the operation of the L3 automated driving system, since there is no need to supervise the vehicle at all times, the driver is often disengaged from the driving task and engages in a variety of non-driving-related tasks (NDRTs). When the autonomous driving system (ADS) encounters an unexpected situation and issues a takeover request (TOR), whether the driver can recover the situation awareness (SA) in time is the key to ensure the safety of the takeover. In this study, the theory of attention resource allocation is introduced to more accurately model the dynamic process of the SA recovery. Moreover, the attention allocation model is further developed, and the affecting factors of attention allocation are quantified. An experiment with 90 participants using driving simulator was conducted for different road scenarios to verify the proposed SA model. The model proposed in this study can accurately predict the SA value of the driver under different road scenarios and the time required for the driver to recover to the maximum SA value, which provides a reference for the scientific design of dynamic takeover lead time. The results also show that as the radius of curvature of the road decreased, the level of SA recovery would become progressively worse.

Language: en

Keywords

Attention distribution; Autonomous driving; Driving simulator; Road alignment; Situation awareness