Journal of Cadastre & Land InformatiX (지적과 국토정보)

LX Spatial Information Research Institute (한국국토정보공사 공간정보연구원)
권호 표지
DOI QR코드

DOI QR Code

A Study on Real-time Autonomous Driving Simulation System Construction based on Digital Twin - Focused on Busan EDC -

디지털트윈 기반 실시간 자율주행 시뮬레이션 시스템 구축 방안 연구 - 부산 EDC 중심으로 -

  • Kim, Min-Soo (Department of Computer Engineering, Daejeon University) ;
  • Park, Jong-Hyun (Department of Computer Engineering, Daejeon University) ;
  • Sim, Min-Seok (Department of Computer Engineering, Daejeon University)
  • 김민수 (대전대학교 컴퓨터공학과) ;
  • 박종현 (대전대학교 컴퓨터공학과) ;
  • 심민석 (대전대학교 컴퓨터공학과)
  • Received : 2023.10.06
  • Accepted : 2023.11.29
  • Published : 2023.12.10
Download PDF
⟨ Previous Next ⟩

Abstract

Recently, there has been a significant interest in the development of autonomous driving simulation environment based on digital twin. In the development of such digital twin-based simulation environment, many researches has been conducted not only performance and functionality validation of autonomous driving, but also generation of virtual training data for deep learning. However, such digital twin-based autonomous driving simulation system has the problem of requiring a significant amount of time and cost for the system development and the data construction. Therefore, in this research, we aim to propose a method for rapidly designing and implementing a digital twin-based autonomous driving simulation system, using only the existing 3D models and high-definition map. Specifically, we propose a method for integrating 3D model of FBX and NGII HD Map for the Busan EDC area into CARLA, and a method for adding and modifying CARLA functions. The results of this research show that it is possible to rapidly design and implement the simulation system at a low cost by using the existing 3D models and NGII HD map. Also, the results show that our system can support various functions such as simulation scenario configuration, user-defined driving, and real-time simulation of traffic light states. We expect that usability of the system will be significantly improved when it is applied to broader geographical area in the future.

최근 자율주행 분야에서 실세계와 유사한 디지털트윈 기반의 가상 시뮬레이션 환경 구축에 대한 관심이 크게 증가하고 있다. 디지털트윈 기반의 시뮬레이션 환경에서 자율주행의 성능 및 기능 검증뿐만 아니라, 딥러닝을 위한 가상 학습데이터 생성 연구도 활발히 수행되고 있다. 그러나, 이러한 디지털트윈 기반 자율주행 시뮬레이션 시스템은 고정밀 데이터 구축과 시스템 개발에 많은 시간과 비용을 필요로 하는 문제를 가지고 있다. 이에 본 연구에서는 기 구축된 3차원 입체모형과 정밀도로지도만을 이용하여 디지털트윈 기반의 실시간 자율주행 시뮬레이션 시스템을 신속히 설계하고 구현하기 위한 방안을 제시하고자 한다. 구체적으로 부산 EDC 지역에 대한 FBX의 3D 입체모형과 NGII HD Map을 CARLA에 통합하는 방법과 CARLA 기능을 추가 및 수정하는 방법을 제시한다. 본 연구 결과는 기존의 3D 입체모형과 NGII HD 맵을 활용하면 저렴한 비용으로 신속한 시뮬레이션 시스템의 설계 및 구현이 가능함을 보여준다. 또한, 시뮬레이션 시나리오 구성, 사용자 맞춤형 주행, 실시간 신호등 상태 시뮬레이션 등의 다양한 기능도 지원할 수 있다. 향후 광범위한 지역에 대하여 시스템이 적용되는 경우에 시스템의 활용도가 크게 향상될 것으로 기대된다.

Keywords

Acknowledgement

이 논문은 2022학년도 대전대학교 교내학술연구비 지원에 의해 연구되었음.

References

  1. Na YS, Kim SK, Kim YS, Park JY, Jeong JM, Jo KC, Lee SG, Cho SG, Sunwoo MH, Oh JM. 2020. HD Map Usability Verification for Autonomous Car. Transactions of the Korean Society of Automotive Engineers. 28(11):797-808. https://doi.org/10.7467/KSAE.2020.28.11.797
  2. Lee MH, Jang IS, Kim MS. 2021. A Study on Data Model Conversion Method for the Application of Autonomous Driving of Various Kinds of HD Map. Journal of Cadastre & Land InformatiX. 51(1):39-51.
  3. Lih BJ. 2023. A Study on Construction & Management of Urban Spatial Information Based on Digital Twin. Journal of Cadastre & Land InformatiX. 53(1): 47-63.
  4. Jeong DW, Choi WG, Hong SH, Cho KJ, Park JH. 2023. Comparative Analysis of Digital Twin Trends in Urban Area. Journal of Cadastre & Land InformatiX. 53(1):123-137.
  5. Amini A, Gilitschenski I, Phillips J, Moseyko J, Banerjee R, Karaman S, Rus D. 2020. Learning Robust Control Policies for End-to-End Autonomous Driving From Data-Driven Simulation. IEEE Robotics and Automation Letters. 5(2):1143-1150. https://doi.org/10.1109/LRA.2020.2966414
  6. Deschaud JE. 2021a. KITTI-CARLA: a KITTI-like Dataset Generated by CARLA Simulator. https://arxiv.org/abs/2109.00892. Last accessed 3 October 2023.
  7. Deschaud JE, Duque D, Richa JP, Velasco-Forero S, Marcotegui B, Goulette F. 2021b. Paris-CARLA-3D: A Real and Synthetic Outdoor Point Cloud Dataset for Challenging Tasks in 3D Mapping. Remote Sensing. 13(22):4713.
  8. Dosovitskiy A, Ros G, Codevilla F, Lopez A, Koltun V. 2017. CARLA: An Open Urban Driving Simulator. In Proceedings of Conference on Robot Learning, 13-15 November 2017, CA, USA. 1-16.
  9. Geiger A, Lenz P, Urtasun R. 2021. Are We Ready for Autonomous Driving? The KITTI Vision Benchmark Suite. In Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 16-21 June 2012, RI, USA. 3354-3361.
  10. Jeon HM, Pham LH, Tran DNN, Nguyen HH, Jeon JW. 2022. 3D Synthetic Image Training and Testing Data for Autonomous Driving. In Proceedings of IEEE International Conference on Consumer Electronics-Asia, 26-28 October 2022, Yeosu, Korea. 1-4.
  11. Kim MS, Jang IS. 2022. Development of System for Collecting User-specified Training Data for Autonomous Driving Based on Virtual Road Environment. Sensors and Materials. 34(12):4813-4825. https://doi.org/10.18494/SAM3966
  12. Liu W, Anguelov D, Erhan D, Szegedy C, Reed S, Fu CY, Berg AC. 2016. SSD: Single Shot MultiBox Detector. Lecture Notes in Computer Science. 9905:21-37.
  13. Motta J, Bellotti F, Berta R, Capello A, Cossu M, Gloria AD, Lazzaroni L, Bonora S. 2022. Developing a Synthetic Dataset for Driving Scenarios. Lecture Notes in Electrical Engineering. 866:310-316 https://doi.org/10.1007/978-3-030-95498-7_43
  14. Niranjan DR, VinayKarthik BC, Mohana. 2021. Deep Learning based Object Detection Model for Autonomous Driving Research using CARLA Simulator. In Proceedings of IEEE International Conference on Smart Electronics and Communication, 7-9 October 2021, Trichy, India. 1251-1258.
  15. Online OpenDRIVE Viewer [Internet]. [https://odrviewer.io/]. Last accessed 5 Nov 2023.
  16. Osinski B, Jakubowski A, Ziecina P, Milos P, Galias C, Homoceanu S, Michalewski H. 2020. Simulation-Based Reinforcement Learning for Real-World Autonomous Driving. In Proceedings of IEEE International Conference on Robotics and Automation, 31 May-31 August 2020, Paris, France, 6411-6418.
  17. Pena J, Bergasa LM, Antunes M, Arango F, Gomez-Huelamo C, Lopez-Guillen E. 2022. AD PerDevKit: An Autonomous Driving Perception Development Kit using CARLA simulator and ROS. In Proceedings of IEEE International Conference on Intelligent Transportation Systems, 8-12 October, Macau, China. 4095-4100.
  18. Rong G. Shin BH, Tabatabaee H, Lu Q, Lemke S, Mozeiko M, Boise E, Uhm G, Gerow M, Mehta S, Agafonov E, Kim TH, Sterner E, Ushiroda K, Reyes M, Zelenkovsky D, Kim S. 2020. LGSVL Simulator: A High Fidelity Simulator for Autonomous Driving. In Proceedings of IEEE International Conference on Intelligent Transportation Systems, 20-23 September 2020, Rhodes, Greece. 1-6.
  19. Shah S, Dey D, Lovett C, Kapoor A. 2018. AirSim: High-Fidelity Visual and Physical Simulation for Autonomous Vehicles. Field and Service Robotics. 5:621-635. https://doi.org/10.1007/978-3-319-67361-5_40
  20. Sun P, Kretzschmar H, Dotiwalla X, Chouard A, Patnaik V, Tsui P, Guo J, Zhou Y, Chai Y, Caine B, Vasudevan V, Han W, Ngiam I, Zhao H, Timofeev A, Ettinger S, Krivokon M, Gao A, Joshi A, Zhang Y, Shlens J, Chen Z, Anguelov D. Scalability in Perception for Autonomous Driving: Waymo Open Dataset. 2020. In Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition, 13-19 June 2020, WA, USA. 2443-2451.
  21. Yao S, Zhang J, Hu Z, Wang Y, Zhou X. 2018. Autonomous-driving vehicle test technology based on virtual reality. The Journal of Engineering. 2018(16):1768-1771.
  22. Dworak D, Ciepiela F, Derbisz J, Izzat I, Komorkiewicz M, Wojcik M. 2019. Performance of LiDAR Object Detection Deep Learning Architectures based on Artificially Generated Point Cloud Data from CARLA Simulator. In Proceedings of IEEE International Conference on Methods and Models in Automation and Robotics, 26-29 August 2019, Miedzyzdroje, Poland. 600-605.