Journal of Auto-vehicle Safety Association (자동차안전학회지)

Korean Auto-vehicle Safety Association (한국자동차모빌리티안전학회)
권호 표지
DOI QR코드

DOI QR Code

Evaluation of Autonomous and Manual Vehicle Interactions Based on Psychological Safety Margin Using Waymo Open Dataset

Waymo Open Dataset을 활용한 심리적 안전 마진 기반 자율차-비자율차 상호작용 평가

  • 최원명 (한양대학교 스마트시티공학과) ;
  • 이호윤 (한국건설기술연구원 도로교통연구본부) ;
  • 오철 (한양대학교 교통물류공학과)
  • Received : 2026.05.11
  • Accepted : 2026.06.05
  • Published : 2026.06.30
Download PDF
⟨ Previous Next ⟩

Abstract

The advancement of autonomous driving technology has heightened the need to ensure not only physical safety but also the psychological safety of human drivers in mixed traffic environments. Existing interaction safety metrics, such as Post-Encroachment Time (PET), provide thresholds for physical conflicts but fail to reflect the psychological safety of manual vehicle drivers interacting with autonomous vehicles. This study defines psychological safety as a state in which a human driver of a manual vehicle does not feel anxiety due to the behavior of an autonomous vehicle. Accordingly, we propose that interaction safety evaluation criteria should complement existing physical safety indicators by explicitly incorporating psychological safety. The proposed methodology is to assess autonomous vehicle-manual vehicle interaction safety based on the Psychological Safety Margin (PSM), defined as the set of preemptive and yielding threshold speeds within which an autonomous vehicle can pass through an intersection whilst ensuring the psychological safety of the human driver. Based on the PSM, a Psychological Safety Zone (PSZ) and a Psychological Risk Zone (PRZ) are further established; when the autonomous vehicle's speed falls within the PSZ, the driver's psychological safety is ensured, whereas a speed within the PRZ indicates it is not. A Psychological Safety Score (PSS) was derived by scoring the proportion of interaction time satisfying the PSZ, and utilized as an interaction safety assessment index. The proposed methodology was applied to the Waymo Open Dataset, and the results quantitatively demonstrate differences in psychological safety levels by intersection types and interaction scenarios. The outcome of this study can contribute to establishing psychological safety evaluation criteria for improving the social acceptability of autonomous driving technology. In addition, it is expected to support the assessment of autonomous vehicle driving performance and the development of speed management strategies accounting for the psychological safety of road users.

Keywords

Acknowledgement

이 논문은 2023년도 정부(경찰청)의 재원으로 과학치안진흥센터의 지원을 받아 수행된 연구임(과제명: 실도로 기반 Lv.4 자율주행차량 운전능력 평가기술 개발/ 과제번호: RS-2023-00238253).

References

  1. Zhao, J., Zhao, W., Deng, B., Wang, Z., Zhang, F., Zheng, W. and Burke, A. F., 2024, "Autonomous Driving System: A Comprehensive Survey," Expert Systems with Applications, Vol. 242, pp. 122836.
  2. 김호선, 조영, 김민경, 오철, 이설영, 이조영, 2025, "Level 3 자율주행차의 자율주행모드와 수동주행모드의 주행안전성 영향요인 비교 분석," 한국ITS학회 논문지, Vol. 24, No. 3, pp. 174~193.
  3. Ma, Z. and Zhang, Y., 2024, "Driver-Automated Vehicle Interaction in Mixed Traffic: Types of Interaction and Drivers' Driving Styles," Human Factors, Vol. 66, No. 2, pp. 544~561.
  4. Albano, G., Mattas, K., Donà, R., Vass, S., Suarez-Bertoa, R., Galassi, M. C. and Ciuffo, B., 2024, "Drivers' Behavior at Unsignalized Intersections: An Empirical Analysis and Derivation of Requirements for the European Regulation 1426/2022 Concerning the Type-Approval of Automated Driving Systems," Data Science for Transportation, Vol. 6, No. 1, pp. 3.
  5. Li, D., Liu, A., Pan, H. and Chen, W., 2023, "Safe, Efficient and Socially-Compatible Decision of Automated Vehicles: A Case Study of Unsignalized Intersection Driving," Automotive Innovation, Vol. 6, No. 2, pp. 281~296.
  6. 조영, 이설영, 오철, 서원호, 김형수, 2022, "교차로 차량 주행안전성 예측 기반 위험상황 검지 기술 개발," 대한교통학 회지, Vol. 40, No. 2, pp. 245~259.
  7. Palatinus, Z., Volosin, M., Csábi, E., Hallgató, E., Hajnal, E., Lukovics, M. and Majó-Petri, Z., 2022, "Physiological Measurements in Social Acceptance of Self Driving Technologies," Scientific Reports, Vol. 12, No. 1, pp. 13312.
  8. Taniguchi, A., Enoch, M., Theofilatos, A. and Ieromonachou, P., 2022, "Understanding Acceptance of Autonomous Vehicles in Japan, UK, and Germany," Urban, Planning and Transport Research, Vol. 10, No. 1, pp. 514~535.
  9. Kim, H., Lee, H., Oh, C. and Yeo, H., 2025, "Insights from Psychophysiological Workload Analysis of HumanDriven Vehicle Drivers in Interactions with Autonomous Vehicles," Accident Analysis & Prevention, Vol. 223, pp. 108252.
  10. Hu, X., Zheng, Z., Chen, D., Zhang, X. and Sun, J., 2022, "Processing, Assessing, and Enhancing the Waymo Autonomous Vehicle Open Dataset for Driving Behavior Research," Transportation Research Part C: Emerging Technologies, Vol. 134, pp. 103490.
  11. Eilers, P. H. and Marx, B. D., 1996, "Flexible Smoothing with B-Splines and Penalties," Statistical Science, Vol. 11, No. 2, pp. 89~121.
  12. Gabaire, M., Ghomi, H. and Hussein, M., 2025, "Investigating the Contributing Factors to Autonomous VehicleRoad User Conflicts: A Data-Driven Approach," Accident Analysis & Prevention, Vol. 211, pp. 107898.
  13. Peesapati, L. N., Hunter, M. P. and Rodgers, M. O., 2018, "Can Post Encroachment Time Substitute Intersection Characteristics in Crash Prediction Models?," Journal of Safety Research, Vol. 66, pp. 205~211.
  14. Paul, M. and Ghosh, I., 2020, "Post Encroachment Time Threshold Identification for Right-Turn Related Crashes at Unsignalized Intersections on Intercity Highways under Mixed Traffic," International Journal of Injury Control and Safety Promotion, Vol. 27, No. 2, pp. 121~135.