Citation

Guo Z, Guo W, Tan J. China Saf. Sci. J. 2022; 32(1): 65-71.

Vernacular Title

驾驶员接管自动驾驶车辆的眼动特性和行为分析

Copyright

(Copyright © 2022, China Occupational Safety and Health Association, Publisher Gai Xue bao)

DOI

10.16265/j.cnki.issn1003-3033.2022.01.009

PMID

unavailable

Abstract

In order to explore visual characteristics of drivers while taking over automated vehicles, relationship between their eye movement and take-over response and manipulation behavior was analyzed, and driving simulation tests were conducted to collect driving behavior and eye movement data. Then, their visual perception characteristics in different take-over scenarios were analyzed by using statistical method, and variance law of eye movement indexes before and after TOR were explored. Finally, driving behavior was analyzed based on visual distribution and pupil change characteristics, and intrinsic relationship between eye movement features and take-over response and driving manipulation strategies was revealed. The results show that compared with static scenarios, drivers perceive elements induced from dynamic scenarios more frequently and the average fixation time is much shorter before TOR. Meanwhile, their visual distribution characteristics have a significant correlation with take-over behavior. After TOR, drivers' fixation time increases, while blinking frequency decreases, pupil diameter dilates and saccade amplitude increases. The pupil difference of drivers in different scenarios indicates that they have better vigilance level and more stable manipulation strategy when dealing with dynamic scenarios.

===

为探究接管自动驾驶车辆期间驾驶员的视觉特性,分析眼动与接管反应操控行为的关系,开展驾驶模拟试验收集驾驶行为及眼动数据。运用统计学方法,分析驾驶员感知不同接管场景的视觉特性,探究接管请求（TOR）前后眼动指标的变化规律;并基于视觉分配和瞳孔变化特性分析驾驶行为,揭示眼动特性与接管反应及驾驶操纵策略的内在联系。结果表明:TOR前,相较于静态场景,驾驶员感知动态场景诱发元素扫视更频繁且平均注视时间更短;此时驾驶员的视觉分配特性与其接管反应行为存在显著相关性。TOR后,驾驶员的注视时间增加,眨眼频率降低,瞳孔直径扩张,眼跳幅度增大;不同场景下驾驶员的瞳孔差异表明其应对动态场景时具备更好的警戒水平和更平稳的操纵策略。

Language: en