DOI QR코드

DOI QR Code

다중센서 기반 차선정보 시공간 융합기법

Lane Information Fusion Scheme using Multiple Lane Sensors

  • 이수목 (서울대학교 전기정보공학부 및 뉴미디어통신공동연구소) ;
  • 박기광 (서울대학교 전기정보공학부 및 뉴미디어통신공동연구소) ;
  • 서승우 (서울대학교 전기정보공학부 및 뉴미디어통신공동연구소)
  • Lee, Soomok (Dept. of Electrical and Computer Engineering, INMC, Seoul National University) ;
  • Park, Gikwang (Dept. of Electrical and Computer Engineering, INMC, Seoul National University) ;
  • Seo, Seung-woo (Dept. of Electrical and Computer Engineering, INMC, Seoul National University)
  • 투고 : 2015.10.19
  • 심사 : 2015.11.19
  • 발행 : 2015.12.25

초록

단일 카메라 센서를 기반으로 한 차선검출 시스템은 급격한 조도 변화, 열악한 기상환경 등에 취약하다. 이러한 단일 센서 시스템의 한계를 극복하기 위한 방안으로 센서 융합을 통해 성능 안정화를 도모할 수 있다. 하지만, 기존 센서 융합의 연구는 대부분 물체 및 차량을 대상으로 한 융합 모델에 국한되어 차용하기 어렵거나, 차선 센서의 다양한 신호 주기 및 인식범위에 대한 상이성을 고려하지 않은 경우가 대부분이었다. 따라서 본 연구에서는 다중센서의 상이성을 고려하여 차선 정보를 최적으로 융합하는 기법을 제안한다. 제안하는 융합 프레임워크는 센서 별 가변적인 신호처리 주기와 인식 신뢰 범위를 고려하므로 다양한 차선 센서 조합으로도 정교한 융합이 가능하다. 또한, 새로운 차선 예측 모델의 제안을 통해 간헐적으로 들어오는 차선정보를 세밀한 차선정보로 정밀하게 예측하여 다중주기 신호를 동기화한다. 조도환경이 열악한 환경에서의 실험과 정량적 평가를 통해, 제안하는 융합 시스템이 기존 단일 센서 대비 인식 성능이 개선됨을 검증한다.

Most of the mono-camera based lane detection systems are fragile on poor illumination conditions. In order to compensate limitations of single sensor utilization, lane information fusion system using multiple lane sensors is an alternative to stabilize performance and guarantee high precision. However, conventional fusion schemes, which only concerns object detection, are inappropriate to apply to the lane information fusion. Even few studies considering lane information fusion have dealt with limited aids on back-up sensor or omitted cases of asynchronous multi-rate and coverage. In this paper, we propose a lane information fusion scheme utilizing multiple lane sensors with different coverage and cycle. The precise lane information fusion is achieved by the proposed fusion framework which considers individual ranging capability and processing time of diverse types of lane sensors. In addition, a novel lane estimation model is proposed to synchronize multi-rate sensors precisely by up-sampling spare lane information signals. Through quantitative vehicle-level experiments with around view monitoring system and frontal camera system, we demonstrate the robustness of the proposed lane fusion scheme.

키워드

참고문헌

  1. E. Dickmanns and D. Mysliwetz, "Recursive 3D road and relative ego-state recognition," IEEE Trans. on Pattern Analysis and Machine Intelligence. Vol. 14. no. 2. pp 119- Feb. 1992.
  2. M. Nieto and L. Salgado, "Robust multiple lane road modeling based on perspective analysis," Image Processing, ICIP, San Diego, 2008.
  3. J. McCall and M. Trivedi "Video-based Lane Estimation and Tracking for Drivier Assistance: Survey, System, and Evaluation," IEEE Trans. on Intelligent Transportation Systems, Vol. 7 No. 1, March 2006.
  4. S. Kang, S. Lee, J. Hur and S. Seo, "Multi-lane Detection based on Accurate Geometric Lane Estimation in Highway Scenarios," in Intelligent Vehicles Symposium(IV), Michigan, 2014.
  5. J. Hur, S. Seo. "Multi-lane detection in urban driving environments using conditional random fields." in Intelligent Vehicles Symposium, Austrailia, 2013.
  6. B. Huval and A. Y. Ng, "An Empirical Evaluation of Deep Learning on Highway Driving", arXiv, April 2015.
  7. S. Lee, S. Kim, S. Seo, "Accurate Ego-Lane Recognition utilizing Multiple Road Characteristics in a Bayesian Network Framework," IEEE Conf. on Intelligent Vehicles Symposium(IV) pp 543-548, Korea, June 2015.
  8. Y. Son, and C. Chung, "Robust Multirate Control Scheme With Predictive Virtual Lanes for Lane-Keeping System of Autonomous Highway Driving," IEEE Trans. on Vehicular Technology. Vol. 64. no. 8. pp 3378-3391 Sep. 2014.
  9. Y. Son, and C. Chung, "Predictive Virtual Lane Method using Relative Motions between a Vehicle and Lanes," International Journal of Control, Automation, and Systems Vol. 13, no. 1, pp 146-155, 2015. https://doi.org/10.1007/s12555-013-0276-5
  10. T. Li and D. Zhidong "A new 3D LIDAR-based lane markings recognition approach," IEEE Int. Conf. on Robotics and Biomimetics (ROBIO), pp 2197-2202, Dec. 2013.
  11. C. Kum, and H. Kim "Development of Lane Detection System using Surrounding View Image of Vehicle," in Conf. on the Korean Society Of Broad Engineers, pp 331-334, Seoul, Korea, June 2013.
  12. M. Aeberhard, and T. Bertram, "Track-to-Track Fusion With Asynchronous Sensors Using Information Matrix Fusion for Surround Environment Perception," IEEE Trans. on Intelligent Transportation System. Vol. 13. no. 4. pp 1717-1726 June. 2013. https://doi.org/10.1109/TITS.2012.2202229
  13. Y. Son and C. Chung, "Asynchronous Sensor Fusion using Multi-rate Kalman Filter," Trans. of the Korean Institute of Electrical Engineers, Vol. 63, No. 11, pp. 1551-1558, 2014. https://doi.org/10.5370/KIEE.2014.63.11.1551
  14. G. Thomaidis and A. Amditis, "Multi-sensor tracking and lane estimation in highly automated vehicles," in IET Intelligent Transportation Systems, Vol 7, no. 1, pp. 160-169, 2013. https://doi.org/10.1049/iet-its.2011.0166
  15. S. Boyd and L. Vandenberghe "Convex Optimization", Cambridge University Press, Ch. 6, pp. 291-342 , 2004.