All Issue

2026 Vol.44, Issue 5 Preview Page

Article

Classification of Urban Road Segments Based on Driving Safety Characteristics of Autonomous and Manual Vehicles

자율차와 비자율차 주행 안전 특성에 기반한 도심 도로 구간 유형화 연구

Hyein LEE1
,
Hyeongyun KI1
,
Jihan KIM2
,
Seolyoung LEE3*
이 혜인1
,
기 현균1
,
김 지한2
,
이 설영3*
1Researcher, Smart Transportation Research Division, The Seoul Institute, Seoul 06756, Korea
2Manager, Research Data Team, The Seoul Institute, Seoul 06756, Korea
3Research Fellow, Smart Transportation Research Division, The Seoul Institute, Seoul 06756, Korea
1서울연구원 스마트교통연구실 연구원
2서울연구원 연구데이터팀 책임행정원
3서울연구원 스마트교통연구실 연구위원
*Corresponding Author
31 October 2026. pp. 871-894
PDF XML Citation
Abstract

The expansion of autonomous vehicle (AV) pilot operation zones has accelerated the emergence of mixed traffic environments in which AVs and manual vehicles (MVs) coexist on urban roads. This study targeted the Seoul Gangnam Autonomous Vehicle Pilot Operation Zone, integrating AV driving records and takeover events with risky driving behavior data of conventional taxis based on digital tachograph (DTG) records. The study aims to classify urban road segments according to their driving safety characteristics and analyze the safety profile of each cluster. Risky driving behavior occurrence rates, AV takeover rates, and average travel speeds at the segment level were used as analysis variables to compare the driving characteristics of the two vehicle types, and a k-means++ clustering analysis was applied to classify 65 road segments into five distinct types. The results indicate that conventional taxis exhibited a relatively even distribution of diverse risky driving behaviors such as sudden acceleration, sudden start, and sudden lane change, whereas AVs showed risky behaviors concentrated in sudden acceleration and sudden deceleration while turn and lane-change type behaviors remained limited, empirically confirming the distinct driving characteristics of MVs and AVs in real-world conditions. The five identified cluster types—Baseline urban traffic, AV-dominant conflict, MV-dominant conflict, Low-risk stable, and Takeover-concentrated—each reflect distinct road environmental conditions. In particular, the AV-dominant conflict and Takeover-concentrated clusters were associated with road environments requiring heightened attention from AVs, such as intersections, protected zones, and construction sections. This study contributes to the literature by providing road segment-level diagnostic evidence for AV-friendly road operation in mixed traffic environments based on real-world AV operation data, offering implications for operational strategies and safety management in pilot operation zones.

Keywords
autonomous vehicle
autonomous vehicle designated pilot operation zone
cluster analysis
risky driving behavior
road segment classification

자율주행 기술의 발전과 시범운행지구 확대로 자율주행 차량과 비자율주행 차량이 혼재하는 도심 교통 환경이 점차 현실화되고 있다. 본 연구는 서울 강남 자율주행자동차 시범운행지구를 대상으로 자율주행 차량의 주행기록 및 제어권 전환(Takeover) 데이터와 일반 택시의 디지털운행기록장치(DTG) 기반 위험운전행동 데이터를 수집하여, 도심 도로 구간을 주행 안전 특성에 따라 유형화하고 분석하였다. 위험운전행동 발생 비율, 제어권 전환 발생 비율, 도로 구간별 평균 통행속도를 변수로 설정하여 차량 유형별 주행 특성을 비교한 후, k-means++ 군집 분석을 통해 65개 도로 구간을 5개 유형으로 분류하였다. 분석 결과, 일반 택시는 급가속･급출발･차로변경 등 다양한 위험운전행동이 고르게 발생한 반면, 자율주행 차량은 급가속･급감속 유형이 상대적으로 두드러지고 회전･차로변경 유형의 행동은 제한적으로 나타났다. 이처럼 인적 운전자와 자율주행 시스템의 상이한 주행 특성이 실도로에서 확인되었다. 도출된 5개 유형(도심 일반 혼잡, 자율차 상충, 비자율차 상충, 저위험 주행 안정, 제어권 전환 집중)은 각기 다른 도로 환경 조건을 반영하며, 특히 자율차 상충 구간과 제어권 전환 집중 구간은 교차로･보호구역･공사구간 등 자율주행 차량의 주의가 요구되는 도로 환경과 연관되는 것으로 나타났다. 본 연구는 실도로 운행 데이터를 기반으로 혼합 교통 환경에서 자율주행 친화적 도로 운영을 위한 도로 구간 단위의 진단 근거를 제시하였다는 점에서 의의가 있다.

키워드
자율주행자동차
자율주행자동차 시범운행지구
군집 분석
위험운전행동
도로 구간 유형화
References
1

Arthur D., Vassilvitskii S. (2007), K-Means++: The Advantages of Careful Seeding, Proceedings of the 18th Annual ACM-SIAM Symposium on Discrete Algorithms, Society for Industrial and Applied Mathematics, 1027-1035.

2

Chen S., Cheng K., Yang J., Zang X., Luo Q., Li J. (2023), Driving Behavior Risk Measurement and Cluster Analysis Driven by Vehicle Trajectory Data, Appl. Sci., 13(9), MDPI, 5675.

10.3390/app13095675
3

de Zepeda M. V. N., Meng F., Su J., Zeng X. J., Wang Q. (2021), Dynamic Clustering Analysis for Driving Styles Identification, Eng. Appl. Artif. Intell., 97, Elsevier, 104096.

10.1016/j.engappai.2020.104096
4

Debbaghi F. Z., Rombaut E., Vanhaverbeke L. (2025), Lessons Learned From Shared Automated Vehicles Pilots in Europe: An Evaluation of Safety, Traffic, and User Acceptance, Case Stud. Transp. Policy, 20, Elsevier, 101447.

10.1016/j.cstp.2025.101447
5

Du X., Kang X., Gao Y., Wang X. (2024), Driving Behavior Characterization and Traffic Emission Analysis Considering the Vehicle Trajectory, Front. Psychol., 14, Frontiers Media, 1341611.

10.3389/fpsyg.2023.134161138348110PMC10860677
6

Guo J., Liu Y., Zhang L., Wang Y. (2018), Driving Behaviour Style Study With a Hybrid Deep Learning Framework Based on GPS Data, Sustainability, 10(7), MDPI, 2351.

10.3390/su10072351
7

He Y., Xiang D., Wang D. (2025), Traffic Safety Evaluation of Emerging Mixed Traffic Flow at Freeway Merging Area Considering Driving Behavior, Sci. Rep., 15(1), Springer Nature, 10686.

10.1038/s41598-025-94658-y40155670PMC11953356
8

Hennig C. (2015), What Are the True Clusters?, Pattern Recognit. Lett., 64, Elsevier, 53-62.

10.1016/j.patrec.2015.04.009
9

Jin J. B., Kim S. H., Kang N. I., Kim H. S., Lee S. H., Noh B. J. (2024), Multi-Dimensional Analytical for Risky Driving Behaviors of Commercial Vehicles Based on Digital TachoGraph Data Using Data Cube Model, J. Korean Soc. Transp., 42(6), Korean Society of Transportation, 627-648.

10.7470/jkst.2024.42.6.627
10

Jo Y., Jung A., Oh C., Park J. (2024), Characterizing the Driving Behavior of Manual Vehicles Following Autonomous Vehicles and Its Impact on Mixed Traffic Performance, Transp. Res. Part F Traffic Psychol. Behav., 107, Elsevier, 69-83.

10.1016/j.trf.2024.08.028
11

Jo Y., Jung A. R., Park H. J., Park J. H., Oh C. (2022), Prioritizing Driving Safety Indicators Using Real-World C-ITS Data to Identify Hazardous Freeway Sections, J. Korean Soc. Transp., 40(6), Korean Society of Transportation, 863-878.

10.7470/jkst.2022.40.6.863
12

Kim H. S., Jo Y., Kim M. K., Oh C., Lee S. Y., Lee J. Y. (2025a), Comparative Analysis of Road Traffic Factors Affecting Autonomous vs. Manual Driving Safety, J. Korean Soc. Intell. Transp. Syst., 24(3), The Korean Society of Intelligent Transport Systems, 174-193.

10.12815/kits.2025.24.3.174
13

Kim M. K., Kim H. S., Jo Y., Oh C., Lee S. Y. (2024), Derivation of Promising Driving Safety Evaluation Indicators for Mixed Traffic Situations in Traffic Simulation Environments, J. Korean Soc. Transp., 42(2), Korean Society of Transportation, 152-167.

10.7470/jkst.2024.42.2.152
14

Kim S. H., Kang K. P., Lee S. J. (2025b), Analysis of Take-Over Cluster Points and Influencing Factors Using Automated Driving Operation Data, J. Korean Soc. Transp., 43(2), Korean Society of Transportation, 145-160.

10.7470/jkst.2025.43.2.145
15

Korea Transportation Safety Authority (2020), Analysis of Correlation Between Dangerous Driving Behavior and Accident Using TIMS Data.

16

Korea Transportation Safety Authority (2025), Digital Tachograph Analysis System (eTAS) User's Guide (한국교통안전공단 운행기록분석시스템 사용자 설명서).

17

Lee E. J., Seo Y. H., Lim D. H., So J. H., Kim H. J. (2024a), Analysis of Urban Hazardous Road for Autonomous Vehicle Using Monitoring Indicator: Focused on Pangyo Zero City, J. Korean Soc. Intell. Transp. Syst., 23(6), The Korean Society of Intelligent Transport Systems, 305-324.

10.12815/kits.2024.23.6.305
18

Lee H. S., Joo J. H., Choo S. H. (2014), A Study on Improving Traffic Safety Evaluation Standard Based on Driving Simulators, The Korea Spatial Planning Review, 80, Korea Research Institute for Human Settlements, 3-15.

10.15793/kspr.2014.80..001
19

Lee S., Jo Y., Kim H., Park J., Oh C. (2024b), Driving Safety Evaluation of Mixed Car-Following Situations by Autonomous and Manual Vehicles at Urban Interrupted Road Facilities, Transp. Res. Rec., 2678(8), National Academy of Sciences: Transportation Research Board, 772-784.

10.1177/03611981231222237
20

Lee S., Kim M., Park J., Oh C. (2025), An Integrated Automated Driving Risk Indicator in Urban Mixed Traffic Environments, Appl. Sci., 15(23), MDPI, 12646.

10.3390/app152312646
21

Liu Y., Li Z., Xiong H., Gao X., Wu J. (2010), Understanding of Internal Clustering Validation Measures, 2010 IEEE International Conference on Data Mining, IEEE, 911-916.

10.1109/ICDM.2010.35
22

Mahmud S. M. S., Ferreira L., Hoque M. S., Tavassoli A. (2017), Application of Proximal Surrogate Indicators for Safety Evaluation: A Review of Recent Developments and Research Needs, IATSS Research, 41(4), International Association of Traffic and Safety Sciences, 153-163.

10.1016/j.iatssr.2017.02.001
23

Negash N. M., Yang J. (2023), Driver Behavior Modeling Toward Autonomous Vehicles: Comprehensive Review, IEEE Access, 11, IEEE, 22788-22821.

10.1109/ACCESS.2023.3249144
24

Oh J. T., Cho J. H., Lee S. Y., Kim Y. S. (2009), Development of a Critical Value According to Dangerous Drive Behaviors, Int. J. Highw. Eng., 11(1), Korean Society of Road Engineers, 69-83.

25

Park E. M., Kang J. H., Bilal M., Lee S. H. (2020), Analysis of Dangerous Driving Behavior by Bus Type Using Digital Tacho Graph in Sejong City, J. Korean Soc. Transp., 38(6), Korean Society of Transportation, 462-472.

10.7470/jkst.2020.38.6.462
26

Seo Y., Lim D., Lee J., So J., Kim H. (2025), A Framework for Evaluating Driving Patterns of Automated Vehicle Based on On-Road Driving Records, IEEE Access, 13, IEEE, 67894-67907.

10.1109/ACCESS.2025.3556299
27

Sohrabi S., Khodadadi A., Mousavi S. M., Dadashova B., Lord D. (2021), Quantifying the Automated Vehicle Safety Performance: A Scoping Review of the Literature, Evaluation of Methods, and Directions for Future Research, Accid. Anal. Prev., 152, Elsevier, 106003.

10.1016/j.aap.2021.106003
28

Yang Z., Yao Y., Zhang L. (2025), Optimizing Autonomous Taxi Deployment for Safety at Skewed Intersections: A Simulation Study, Sensors, 25(11), MDPI, 3544.

10.3390/s2511354440969095PMC12158363
29

Younes L. (2024), Introduction to Machine Learning, arXiv:2409.02668.

30

Zahid M., Chen Y., Khan S., Jamal A., Ijaz M., Ahmed T. (2020), Predicting Risky and Aggressive Driving Behavior Among Taxi Drivers: Do Spatio-Temporal Attributes Matter?, Int. J. Environ. Res. Public Health, 17(11), MDPI, 3937.

10.3390/ijerph1711393732498347PMC7312618
Information
  • Publisher :Korean Society of Transportation
  • Publisher(Ko) :대한교통학회
  • Journal Title :Journal of Korean Society of Transportation
  • Journal Title(Ko) :대한교통학회지
  • Volume : 44
  • No :5
  • Pages :871-894
  • Received Date : 2026-05-04
  • Revised Date : 2026-05-29
  • Accepted Date : 2026-07-21
  • DOI :https://doi.org/10.7470/jkst.2026.44.5.871
Journal Informaiton Journal of Korean Society of Transportation Journal of Korean Society of Transportation
Archives
: 2015 Vol.34~
  • scopus
  • NRF
  • KOFST
  • crossref cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close