Journal of IKEEE (전기전자학회논문지)

Institute of Korean Electrical and Electronics Engineers (한국전기전자학회)
권호 표지
DOI QR코드

DOI QR Code

Interactive ADAS development and verification framework based on 3D car simulator

3D 자동차 시뮬레이터 기반 상호작용형 ADAS 개발 및 검증 프레임워크

  • Cho, Deun-Sol (Dept. of Computer Science and Engineering, Koreatech University) ;
  • Jung, Sei-Youl (Dept. of Computer Science and Engineering, Koreatech University) ;
  • Kim, Hyeong-Su (Dept. of Computer Science and Engineering, Koreatech University) ;
  • Lee, Seung-gi (Dept. of Computer Science and Engineering, Koreatech University) ;
  • Kim, Won-Tae (Dept. of Computer Science and Engineering, Koreatech University)
  • Received : 2018.12.05
  • Accepted : 2018.12.13
  • Published : 2018.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

The autonomous vehicle is based on an advanced driver assistance system (ADAS) consisting of a sensor that collects information about the surrounding environment and a control module that determines the measured data. As interest in autonomous navigation technology grows recently, an easy development framework for ADAS beginners and learners is needed. However, existing development and verification methods are based on high performance vehicle simulator, which has drawbacks such as complexity of verification method and high cost. Also, most of the schemes do not provide the sensing data required by the ADAS directly from the simulator, which limits verification reliability. In this paper, we present an interactive ADAS development and verification framework using a 3D vehicle simulator that overcomes the problems of existing methods. ADAS with image recognition based artificial intelligence was implemented as a virtual sensor in a 3D car simulator, and autonomous driving verification was performed in real scenarios.

자율 주행 차량은 주변 환경의 정보를 수집하는 센서, 측정된 데이터를 판단하는 제어 모듈로 구성된 첨단 운전자 지원 시스템(ADAS)을 기반하고 있다. 최근에 자율주행 기술에 대한 관심이 증가함에 따라 ADAS 입문 개발자들 및 학습자들을 위한 손쉬운 개발프레임워크가 필요하다. 그러나, 기존 개발 및 검증 방식은 고성능 자동차 시뮬레이터를 기반하기 때문에 검증 방법의 복잡성 및 고비용 등의 단점이 있다. 또한, 대부분의 방식은 시뮬레이터로부터 ADAS에서 필요로 하는 센싱 데이터를 직접 제공하지 않으므로 검증 신뢰성의 한계가 있다. 본 논문에서는 기존 방식들의 문제점들을 극복하는 3D 자동차 시뮬레이터를 활용한 상호작용형 ADAS 개발 및 검증 프레임워크를 제시한다. 영상인지 기반의 인공지능을 적용한 ADAS를 3D 자동차 시뮬레이터에서의 가상센서로 구현하고, 실제 시나리오에 자율주행 검증을 진행하였다.

Keywords

JGGJB@_2018_v22n4_970_f0001.png 이미지

Fig. 1. The proposed ADAS development and verification framework. 그림 1. 제안하는 ADAS 개발 및 검증 프레임워크

JGGJB@_2018_v22n4_970_f0002.png 이미지

Fig. 2. The data flow diagram of ADAS development and validation framework. 그림 2. ADAS 개발 및 검증 프레임워크의 데이터 흐름도

JGGJB@_2018_v22n4_970_f0003.png 이미지

Fig. 3. ADAS development and verification configuration based on proposed framework. 그림 3. 제안 프레임워크 기반 ADAS 개발 및 검증 구성

JGGJB@_2018_v22n4_970_f0004.png 이미지

Fig. 4. Object recognition rate verification result. 그림 4. 객체 인식률 검증 결과

JGGJB@_2018_v22n4_970_f0005.png 이미지

Fig. 5. Object recognition in AEB. 그림 5. AEB에서의 객체 인식 사진

JGGJB@_2018_v22n4_970_f0006.png 이미지

Fig. 6. ROI of LKAS and lane recognition during driving. 그림 6. LKAS의 ROI 및 주행 중 차선 인식 사진

JGGJB@_2018_v22n4_970_f0007.png 이미지

Fig. 7. Data including intersections(1st~4th). 그림 7. 교차로를 포함하는 데이터(1회~4회)

JGGJB@_2018_v22n4_970_f0008.png 이미지

Fig. 8. Data not including intersections(1st~4th). 그림 8. 교차로를 포함하지 않는 데이터(1회~4회)

References

  1. M. Konig, and L. Neumayr, "Users' resistance towards radical innovations: The case of the self-driving car," Transportation Research Part F: Traffic Psychology and Behaviour, vol.44, pp. 42-52, 2017. DOI:10.1016/j.trf.2016.10.013
  2. U. Lee, S. Yoon, H. Shim, P. Vasseur, and C. Demonceaux, "Local path planning in a complex environment for self-driving car," The 4th Annual IEEE International Conference on Cyber Technology in Automation, Control and Intelligent, pp.445-450, 2014. DOI:10.1109/CYBER.2014.6917505
  3. J. Wei, J. M. Snider, J. Kim, J. M. Dolan, R. Rajkumar, and B. Litkouhi, "Towards a viable autonomous driving research platform," 2013 IEEE Intelligent Vehicles Symposium, pp.763-770, 2013. DOI:10.1109/IVS.2013.6629559
  4. J. K. Choi, K. Kim, D. Kim, H. Choi, and B. Jang, "Driver-adaptive vehicle interaction system for the advanced digital cockpit," 2018 20th International Conference on Advanced Communication Technology (ICACT), pp.307-310, 2018. DOI:10.23919/ICACT.2018.8323736
  5. Hassan, B, Berssenbrugge, J, Al Qaisi, I, & Stocklein, J, "Reconfigurable driving simulator for testing and training of advanced driver assistance systems," IEEE International Symposium, pp. 337-339 2013. DOI:10.1109/ISAM.2013.6643472
  6. S. Shah, D. Dey, C. Lovett, and A. Kapoor, "Airsim : High-fidelity visual and physical simulation for autonomous vehicles," Field and service robotics, Springer, Cham, pp.621-635.
  7. J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, "You only look once: Unified, real-time object detectio," In Proceedings of the IEEE conference on computer vision and pattern recognition, pp.779-788, 2016.
  8. S. U. Park and J. C. Kim. "Open Source Based Vehicle Control Algorithm Competition Platform for Education," Korea Information Science Society, pp.1411-1413, 2014.
  9. D. S. Choi, T. H. An, K. H. An, and J. D. Choi, "End-to-end learning using TORCS for autonomous driving," The Institute of Electronics and Information Engineers, pp.740-743, 2017.
  10. C. W. Park, S. O. Kim, H. C. Lee, "A Study on Approaching Mode of Adaptive Cruise Control System," The Korean Society Of Automotive Engineers, pp.666-669, 2014. DOI:10.1109/ISAM.2013.6643472
  11. Hassan, B, Berssenbrugge, J, Al Qaisi, I, & Stocklein, J, "Reconfigurable driving simulator for testing and training of advanced driver assistance systems," IEEE International Symposium, pp. 337-339, 2013. DOI:10.1109/ISAM.2013.6643472
  12. J. S. Kim, "Effective Road Distance Estimation Using a Vehicle-attached Black Box Camera," Korea Institute of Information and Communication Engineering, pp.651-658, 2015. DOI:10.6109/jkiice.2015.19.3.651