Analysis on eye movement characteristics and behavior of drivers taking over automated vehicles

doi:10.16265/j.cnki.issn1003-3033.2022.01.009

Abstract

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.

Key words: drivers, take-over request(TOR), automated vehicles, eye movement characteristics, driving behavior

GUO Zihui, GUO Weiwei, TAN Jiyuan. Analysis on eye movement characteristics and behavior of drivers taking over automated vehicles[J]. China Safety Science Journal, 2022, 32(1): 65-71.

Figures/Tables 7

Fig.1

Fig.2

Fig.3

Tab.1

Tab.2

Fig.4

Fig.5

References 20

[1]	陈清泰. 未来汽车如何与交通、城市协同发展[J]. 汽车纵横, 2020(12):22-23.
[2]	边明远, 李克强. 高级别自动驾驶的技术特征和落地场景[J]. 智能网联汽车, 2019(2):23-27.
[3]	SAE International. Surface vehicle recommended practice: taxonomy and definitions for terms related to driving automation systems for on-road motor vehicles[R], 2018.
[4]	乔建刚, 郭蕊, 李维东. 穿城镇道路驾驶员眼动特性分析[J]. 中国安全科学学报, 2020, 30(7):13-18. doi: 10.16265/j.cnki.issn1003-3033.2020.07.003
	QIAO Jiangang, GUO Rui, LI Weidong. Eye movement characteristics of drivers while crossing urban highways[J]. China Safety Science Journal, 2020, 30(7):13-18.
[5]	汪磊, 高杉. 飞行冒险行为的眼动和生理特征研究[J]. 中国安全科学学报, 2020, 30(9):22-28.
	WANG Lei, GAO Shan. Study on eye movement and physiological characteristics of flying risk-taking behaviors[J]. China Safety Science Journal, 2020, 30(9):22-28.
[6]	VICTOR T W, HARBLUK J L, ENGSTROM J A. Sensitivity of eye movement measures to in-vehicle task difficulty[J]. Transportation Research Part F: Traffic Psychology and Behaviour, 2005, 8(2):167-190. doi: 10.1016/j.trf.2005.04.014
[7]	MERAT N, JAMSON A H, LAI F C H, et al. Highly automated driving, secondary task performance, and driver state[J]. Human Factors the Journal of the Human Factors and Ergonomics Society, 2012, 54(5):762-771. doi: 10.1177/0018720812442087
[8]	SCHWALM M, KEINATH A, ZIMMER H D. Pupillometry as a method for measuring mental workload within a simulated driving task[C]. Proceedings of the Annual Meeting of Human Factors and Ergonomics Society-Europe Chapter, 2008: 1-13.
[9]	ZEEB K, BUCHNER A, SCHRAUF M. What determines the take-over time? an integrated model approach of driver take-over after automated driving[J]. Accident Analysis and Prevention, 2015, 78:212-221. doi: 10.1016/j.aap.2015.02.023
[10]	WU Yanbin, KIHARA K, TAKEDA Y, et al. Effects of scheduled manual driving on drowsiness and response to take over request: a simulator study towards understanding drivers in automated driving[J]. Accident Analysis and Prevention, 2019, 124:202-209. doi: 10.1016/j.aap.2019.01.013
[11]	ZHOU Fang, ALSAID A, BLOMMER M, et al. Driver fatigue transition prediction in highly automated driving using physiological features[J]. Expert Systems with Applications, 2020, 147:DOI:10.1016/j. eswa.2020.113204.
[12]	LOUW T, KUO J, ROMANO R, et al. Engaging in NDRTs affects drivers' responses and glance patterns after silent automation failures[J]. Transportation Research Part F: Traffic Psychology and Behaviour, 2019, 62:870-882. doi: 10.1016/j.trf.2019.03.020
[13]	YOON S H, JI Yonggu. Non-driving-related tasks, workload, and takeover performance in highly automated driving contexts[J]. Transportation Research Part F: Psychology and Behaviour, 2019, 60:620-631. doi: 10.1016/j.trf.2018.11.015
[14]	WINTER J C F D, HAPPEE R, MARTENS M H, et al. Effects of adaptive cruise control and highly automated driving on workload and situation awareness: a review of the empirical evidence[J]. Transportation Research Part F: Psychology and Behaviour, 2014, 27:196-217. doi: 10.1016/j.trf.2014.06.016
[15]	MERAT N, JAMSON A H, LAI F C H, et al. Transition to manual: driver behaviour when resuming control from a highly automated vehicle[J]. Transportation Research Part F: Traffic Psychology and Behaviour, 2014, 27:274-282. doi: 10.1016/j.trf.2014.09.005
[16]	GROH K, KUEHBECK T, FLEISCHMANN B, et al. Towards a scenario-based assessment method for highly automated driving functions[C]. Proceedings of the 8^th Conference on Driver Assistance, 2017:1-7.
[17]	GOLD C, DAMBOCK D, LORENZ L, et al. "Take over!" how long does it take to get the driver back into the loop?[C]. Proceedings of the Human Factors and Ergonomics Society 57^th Annual Meeting, 2013:1938-1942.
[18]	殷莉, 吴玲, 路巧珍, 等. 基于扫视特征的车载信息系统与驾驶安全关系研究[J]. 中国安全科学学报, 2015, 25(6):124-128.
	YIN Li, WU ling, LU Qiaozhen, et al. Research on relationship between in-vehicle information system and driving safety based on scanning features[J]. China Safety Science Journal, 2015, 25(6):124-128.
[19]	焦方通, 杜志刚, 王首硕, 等. 城市水下特长隧道弯道驾驶人扫视行为研究[J]. 中国安全科学学报, 2019, 29(7):104-109. doi: 10.16265/j.cnki.issn1003-3033.2019.07.017
	JIAO Fangtong, DU Zhigang, WANG Shoushuo, et al. Research on saccade behavior of drivers in extra-long urban underwater tunnel curves[J]. China Safety Science Journal, 2019, 29(7):104-109. doi: 10.16265/j.cnki.issn1003-3033.2019.07.017
[20]	商艳, 朱守林, 戚春华, 等. 草原公路景观要素影响驾驶员眼动指标研究[J]. 中国安全科学学报, 2017, 27(5):19-24. doi: 10.16265/j.cnki.issn1003-3033.2017.05.004
	SHANG Yan, ZHU Shoulin, QI Chunhua, et al. Impacts of grassland highway landscape elements on driver's eye movement indices[J]. China Safety Science Journal, 2017, 27(5):19-24. doi: 10.16265/j.cnki.issn1003-3033.2017.05.004

眼动指标	均值(标准差)		t值(显著性)	95%置信区间
眼动指标	TOR前5 s	TOR后5 s	t值(显著性)	95%置信区间
注视持续时间/s	3.42(0.91)	4.14(0.82)	6.167^***	(0.53, 0.91)
扫视持续时间/s	0.51(0.25)	0.44(0.16)	-1.266	(-0.04, 0.17)
眨眼频率/(次·s^-1)	0.56(0.34)	0.37(0.21)	-3.486^**	(-1.02, -0.26)
瞳孔直径/mm	3.54(0.75)	3.89(0.72)	5.670^***	(0.21, 0.43)
眼跳幅度/(°)	2.65(0.45)	4.22(0.81)	7.262^***	(0.88, 1.51)

指标		接管反应时间	制动反应时间	转向反应时间
扫视持续时间	相关系数	-0.340^**	-0.240^*	0.419^***
扫视持续时间	显著性	0.004	0.028	<0.001
注视持续时间	相关系数	0.191	0.146	-0.261^*
注视持续时间	显著性	0.069	0.165	0.016