Research on the steady-state of driver takeover in scenarios involving livestock obstructing grassland highways

doi:10.3969/j.issn.1674-8484.2026.03.011

Abstract

Abstract:

A livestock occupation scenario was constructed using driving simulation technology to investigate the impact of different takeover alert schemes on the takeover steady-state performance in specific dynamic scenarios of grassland highways. 4 takeover alert schemes were established: visual alert (VA), visual combined with tactile alert (VTA), visual combined with auditory alert (VAA), visual combined with both auditory and tactile alert (VATA). The entire driving takeover process was divided into initialy takeover phase and later takeover phase based on temporal dimensions, using survival probability model and data envelopment analysis (DEA), in conjunction with fundamental indicators to assess the effectiveness of takeover steady-state transitions. The results indicate that VTA and VATA demonstrates superior performance in terms of survival time, the takeover response is most rapid in the VTA, with a takeover response time of 0.52 s; in the initial takeover phase, integrating auditory alerts can enhance the driver's cognitive-homeostasis transition by 7.2%; in the later takeover phase, the synergistic interaction between tactile and auditory significantly enhances the steady-state transformation benefits, the integration of visual with other modalities contributes to improved steady-state performance, specifically, the steady-state transformation benefits of the VATA is 13.9% higher than alternative schemes.

Key words: grassland highways, takeover steady-state, survival analysis, data envelopment analysis (DEA)

CLC Number:

U471.15

WANG Haixiao, ZHANG Hang, GAO Mingxing, JI Tianxing, WANG Qingju. Research on the steady-state of driver takeover in scenarios involving livestock obstructing grassland highways[J]. Journal of Automotive Safety and Energy, 2026, 17(3): 388-396.

Figures/Tables 12

References 17

[1]	内蒙古自治区统计局. 内蒙古统计年鉴[M]. 北京: 中国统计出版社, 2023: 209.
	Inner Mongolia Autonomous Region Statistics Bureau. Statistical Yearbook of Inner Mongolia[M]. Biejing: China Statistics Press, 2023: 209. (in Chinese)
[2]	王海晓, 丁旭, 郭敏, 等. 基于视觉信息加工的草原公路行车安全性分析[J]. 重庆交通大学学报(自然科学版), 2024, 43(2): 21-25.
	WANG Haixiao, DING Xu, GUO Min, et al. Analysis of driving safety on prairie highway based onvisual information processing[J]. J Chongqing Jiaotong Univ: Nat Sci, 2024, 43(2): 21-25. (in Chinese)
[3]	LYU Zhen, QI Chunhua, ZHU Shoulin. Driver’s attention allocation and mental workload at different random hazard points on prairie highway[J]. J Advan Transport, 2022, 2022: 1-16.
[4]	张亚勤, 李震宇, 尚国斌, 等. 面向自动驾驶的车路云一体化框架[J]. 汽车安全与节能学报, 2023, 14(3): 249-273.
	ZHANG Yaqin, LI Zhenyu, SHANG Guobin, et al. A unified framework for vehicle-infrastructure-cloud autonomous driving[J]. J Autom Safe Energ, 2023, 14(3): 249-273. (in Chinese)
[5]	杨澜, 赵祥模, 王润民, 等. 自动驾驶认知能力测试评价研究综述[J]. 汽车安全与节能学报, 2025, 16(1): 1-15.
	YANG Lan, ZHAO Xiangmo, WANG Runmin, et al. Review on testing and evaluation of cognitive abilities for autonomous vehicles[J]. J Autom Safe Energ, 2025, 16(1): 1-15. (in Chinese)
[6]	张艺竞, 常若松, 马锦飞, 等. L3等级自动驾驶条件下驾驶人接管过程及心理模型的构建[J]. 心理科学, 2019,42(2): 415-421.
	ZHANG Yijing, CHANG Ruosong, MA Jinfei, et al. Take-over process and the construction of drivers’psychological model under level 3 automated driving conditions[J]. Psych Sci, 2019, 42(2): 415-421. (in Chinese)
[7]	Wickens C D. Multiple resources and performance prediction[J]. Theor Issu Ergon Sci, 2002, 3(2): 159-177.
[8]	YUN Hanna, YANG Ji hyun. Multimodal warning design for take-over request in conditionally automated driving[J]. Europ Transport Res Rev, 2020, 12: 1-11. doi: 10.1186/s12544-019-0391-3
[9]	GAO Ya, FENG Zhongxiang, ZHU Dianchen, et al. Analysis of drivers’take-over ability improvement and behavioral steady state in human-machine codriving vehicles[J]. Transport Res Part F: Traf Psych Behav, 2024, 103: 554-573. doi: 10.1016/j.trf.2024.05.007 URL
[10]	Melnicuk V, Thompson S, Jennings P, et al. Effect of cognitive load on drivers’state and task performance during automated driving: Introducing a novel method for determining stabilisation time following take-over of control[J]. Accid Anal Prev, 2021, 151: 105967. doi: 10.1016/j.aap.2020.105967 URL
[11]	王海晓, 梁瑞泽, 高明星, 等. 草原公路驾驶接管风险的模糊推理评估方法研究[J]. 安全与环境学报, 2024, 24(11): 4130-4142.
	WANG Haixiao, LIANG Ruize, GAO Mingxing, et al. Research on fuzzy inference evaluation method for take-over risk in driving on grassland highway[J]. J Safe Environ, 2024, 24(11): 4130-4142. (in Chinese)
[12]	Schneider A, Sonderegger A, Kruegere, et al. The interplay between task difficulty and microsaccade rate: Evidence for the critical role of visual load[J]. J Eye Movem Res, 2021, 13(5): 1-12.
[13]	DING Xu, DU Zhigang, GAO Mingxing, et al. Assessment of takeover safety in human-machine co-driving on grassland highways: Coupling effects of non-driving-related tasks and takeover scenarios[J]. Transport Res Part F: Traf Psych Behav, 2025, 114: 376-408. doi: 10.1016/j.trf.2025.06.003 URL
[14]	WU Yanbin, Kihara K, Takeda Y, et al. Eye movements predict driver reaction time to takeover request in automated driving: A real-vehicle study[J]. Transport Res Part F: Traf Psych Behav, 2021, 81: 355-363. doi: 10.1016/j.trf.2021.06.017 URL
[15]	Kalbfleisch J D, Prentice R L. The Statistical Analysis Of Failure Time Data[M]. Hoboken, NJ: John Wiley & Sons, 2002: 6-10.
[16]	Charnes A, Cooper W W, Rhodes E. Measuring the efficiency of decision making units[J]. Europ J Operat Res, 1978, 2(6): 429-444.
[17]	Omrani H, Amini M, Alizadeh A. An integrated group best-worst method-Data envelopment analysis approach for evaluating road safety: A case of Iran[J]. Measurement, 2020, 152: 1-15.

阶段	模态组合	中位生存时间/ s	95%置信区间/ s
接管初期	VA	1.880	1.78~2.04
	VAA	1.680	1.60~1.74
	VATA	1.500	1.44~1.62
	VTA	1.560	1.44~1.62
接管后期	VA	11.650	11.36~11.95
	VAA	10.730	10.50~11.31
	VTA	11.210	10.72~11.47
	VATA	10.820	10.58~11.22

阶段	模态组合	中位生存时间/ s	95%置信区间/ s
接管初期	VA	1.880	1.78~2.04
	VAA	1.680	1.60~1.74
	VATA	1.500	1.44~1.62
	VTA	1.560	1.44~1.62
接管后期	VA	11.650	11.36~11.95
	VAA	10.730	10.50~11.31
	VTA	11.210	10.72~11.47
	VATA	10.820	10.58~11.22

阶段	项	χ²	df	p值
接管初期	Log Rank检验	31.08	3	0.000
接管初期	Gehan-Breslow-Wilcoxon检验	31.13	3	0.000
接管后期	Log Rank检验	11.23	3	0.011
接管后期	Gehan-Breslow-Wilcoxon检验	19.16	3	0.001

阶段	项	χ²	df	p值
接管初期	Log Rank检验	31.08	3	0.000
接管初期	Gehan-Breslow-Wilcoxon检验	31.13	3	0.000
接管后期	Log Rank检验	11.23	3	0.011
接管后期	Gehan-Breslow-Wilcoxon检验	19.16	3	0.001

阶段	配对组	χ²	p值
接管初期	VAA-VATA	1.263	0.261
	VAA-VA	15.397	0.000
	VAA-VTA	1.730	0.188
	VATA-VA	22.507	0.000
	VATA-VTA	0.042	0.838
	VA-VTA	22.516	0.000
接管后期	VAA-VATA	0.024	0.877
	VAA-VA	8.871	0.003
	VAA-VTA	0.680	0.409
	VATA-VA	9.574	0.002
	VATA-VTA	1.033	0.310
	VA-VTA	4.033	0.045