DOI QR코드

DOI QR Code

교통소통정보 고려 모빌리티 기반 수요응답형 자율배송 서비스 전역경로 생성 시스템 개발

A Study of Global Path Planning System for Traffic Information Aware On-demand Delivery Services Using Autonomous Mobilities

  • 박채훈 (한국자동차연구원 자율주행기술연구소 ) ;
  • 전상윤 (한국자동차연구원 자율주행기술연구소 )
  • Chaehun Park (Automatic Driving Technology Research Division, KATECH) ;
  • Sang-Yun Jeon (Automatic Driving Technology Research Division, KATECH)
  • 투고 : 2024.08.19
  • 심사 : 2024.10.02
  • 발행 : 2024.10.31

초록

자율주행 기술은 기초적인 연구 단계를 넘어 상용화 초입 단계에 접어들었으며, 최근에는 자율주행 기술과 정보통신 기술 기반 지능형교통시스템을 접목한 모빌리티 서비스들이 활발히 개발되고 있다. 본 연구는 모빌리티 기반 서비스 중 하나인 수요응답형 자율배송 서비스의 운영 효율성 향상을 위해 다중 모빌리티 전역경로를 생성하는 것으로, 지능형교통시스템을 통해 수집한 교통소통정보와 서비스 사용자 수요를 고려하여 최단 시간 내에 자율배송을 완료할 수 있는 혼합 정수 최적화 기반 전역경로 생성 시스템이 개발되었다. 개발된 전역경로 생성 시스템은 교통소통정보 갱신 또는 서비스 사용자의 추가 수요 발생에 따라 전역경로를 갱신하며 상암 자율주행 테스트베드의 교통소통정보를 활용하여 수요응답형 자율배송 서비스 운영이 가능함이 확인되었다. 또한, 기존 유인 배송 서비스와 비교분석을 통해 운영 비용 절감과 물품 배송 및 공차 시간이 단축이 확인되었다.

Autonomous driving technologies have entered the initial stage of commercialization. Recently, mobility services that combine autonomous driving technologies and information and communication technologies based intelligent transportation systems are being actively developed. This study develops a global path planning system that considers traffic information and user demands to generate the shortest time paths for autonomous delivery services using Mixed Integer Programming. While providing the autonomous delivery services, the generated paths are updated recursively according to traffic information updates or additional demands. The developed global path planning system was verified by simulations with traffic information in the Sangam autonomous driving test-bed, and comparative analysis with existing manned delivery services shows that operating costs, product delivery time, and empty driving time were reduced.

키워드

과제정보

이 연구는 2024년도 산업통상자원부 및 산업기술기획평가원(KEIT) 연구비 지원에 의한 연구(20024368)입니다.

참고문헌

  1. Alvarenga, G. B., Mateus, G. R. and De Tomi, G.(2007), "A genetic and set partitioning two-phase approach for the vehicle routing problem with time windows", Computers & Operations Research, vol. 34, no. 6, pp.1561-1584.
  2. Helsgaun, K.(2000), "An effective implementation of the Lin-Kernighan traveling salesman heuristic", European Journal of Operational Research, vol. 126, no. 1, pp.106-130.
  3. Helsgaun, K.(2009), "General k-opt submoves for the Lin-Kernighan TSP heuristic", Mathematical Programming Computation, vol. 1, no. 2-3, pp.119-163.
  4. Ho, S. C., Szeto, W. Y., Kuo, Y. H., Leung, J. M., Petering, M. and Tou, T. W.(2018), "A survey of dial-a-ride problems: Literature review and recent developments", Transportation Research Part B: Methodological, vol. 111, pp.395-421.
  5. Janinhoff, L., Klein, R., Sailer, D. and Schoppa, J. M.(2024), "Out-of-home delivery in last-mile logistics: A review", Computers and Operations Research, vol. 168, pp.1-20.
  6. Kim, J., Kim, J. and Yeo, H.(2020), "Study on multi-vehicle routing problem using clustering method for demand responsive transit", Journal of the Korea Institute of Intelligent Transportation Systems, vol. 19, no. 5, pp.82-96.
  7. Kim, K. and Cho, S.(2023), "An optimal route algorithm for automated vehicle in monitoring road infrastructure", Journal of the Korea Institute of Intelligent Transportation Systems, vol. 22, no. 1, pp.265-275.
  8. Laporte, G.(2009), "Fifty years of vehicle routing", Transportation Science, vol. 43, no. 4, pp.408-416.
  9. Mokhtarian, A., Kampmann, A., Lueer, M., Kowalewski, S. and Alrifaee, B.(2021), "A cloud architecture for networked and autonomous vehicles", IFAC-PapersOnLine, vol. 54, no. 2, pp.233-239.
  10. Nazari, M., Oroojlooy, A., Takac, M. and Snyder, L. V.(2018), "Reinforcement learning for solving the vehicle routing problem", 32nd Conference on Neural Information Processing Systems (NeurIPS 2018), pp.1-11.
  11. Oh, S., Sechadri, R., Azevedo, C. L., Kumar, N., Basak, K. and Ben-Akiva, M.(2020), "Assessing the impacts of automated mobility-on-demand through agent-based simulation: A study of Singapore", Transportation Research Part A: Policy Practice, vol. 138, pp.367-388.
  12. Psaraftis, H. N.(1980), "A dynamic programming solution to the single vehicle many-to-many immediate request dial-a-ride problem", Transportation Science, vol. 14, no. 2, pp.130-154.
  13. Toth, P. and Vigo, D.(2014), Vehicle routing: Problems, methods, and applications, SIAM.
  14. Zong, Z., Tong, X., Zheng, M. and Li, Y.(2024), "Reinforcement learning for solving multiple vehicle routing problem with time window", ACM Transactions on Intelligent Systems and Technology, vol. 15, no. 2, pp.1-19.