Journal of the Korean Institute of Intelligent Systems (한국지능시스템학회논문지)

Korean Institute of Intelligent Systems (한국지능시스템학회)
권호 표지
DOI QR코드

DOI QR Code

Novel Collision Warning System using Neural Networks

신경회로망을 이용한 새로운 충돌 경고 시스템

  • Kim, Beomseong (School of Electrical & Electronics Engineering, Yonsei University) ;
  • Choi, Baehoon (School of Electrical & Electronics Engineering, Yonsei University) ;
  • An, Jhonghyun (School of Electrical & Electronics Engineering, Yonsei University) ;
  • Hwang, Jaeho (Advanced Research Team, Hyundai Mobis Co., Ltd.) ;
  • Kim, Euntai (School of Electrical & Electronics Engineering, Yonsei University)
  • Received : 2014.04.22
  • Accepted : 2014.04.28
  • Published : 2014.08.25
Download PDF
⟨ Previous Next ⟩

Abstract

Recently, there are many researches on active safety system of intelligent vehicle. To reduce the probability of collision caused by driver's inattention and mistakes, the active safety system gives warning or controls the vehicle toward avoiding collision. For the purpose, it is necessary to recognize and analyze circumstances around. In this paper, we will treat the problem about collision risk assessment. In general, it is difficult to calculate the collision risk before it happens. To consider the uncertainty of the situation, Monte Carlo simulation can be employed. However it takes long computation time and is not suitable for practice. In this paper, we apply neural networks to solve this problem. It efficiently computes the unseen data by training the results of Monte Carlo simulation. Furthermore, we propose the features affects the performance of the assessment. The proposed algorithm is verified by applications in various crash scenarios.

최근 지능형 자동차의 능동적 안전 기술에 많은 관심이 집중되고 있다. 능동적 안전 기술은 시스템이 운전자에게 위험 상황을 경고 하거나 교통사고를 회피하는 방향으로 차량 제어의 일부를 도와주어 운전자의 부주의 및 과실로 발생하는 교통사고의 발생 가능성을 감소시킨다. 이러한 능동적 안전 기술을 구현하기 위해서는 주변의 환경 정보를 정확하게 인식하고 분석하는 것이 필수적인데, 주변의 차량과 자차량간의 충돌 위험도를 판단하는 것도 그중 하나이다. 하지만 충돌이 발생하기 이전에 충돌 가능성을 판단하는 것은 일반적인 방법으로는 불가능하기 때문에 몬테 카를로 모의실험을 통하여 이를 해결한다. 하지만 몬테 카를로 모의실험은 연산시간이 길기 때문에 실시간으로 동작해야하는 충돌 경고 시스템에는 적합하지 않다. 이때 신경회로망을 이용하면 이러한 문제를 해결 할 수 있으며, 본 논문에서는 자차량과 주변 차량간의 상태정보에서 충돌에 영향을 주는 충돌 특징점을 추출하여 신경회로망의 성능을 높이는 방법을 제안한다. 제안된 알고리즘은 시나리오 기반으로 Tass사의 PreScan을 이용하여 생성된 충돌 실험데이터에 적용하여 성능을 평가한다.

Keywords

References

  1. National Highway Traffic Safety Administration (www.nhtsa.gov)
  2. European New Car Assessment Programme (www.euroncap.com)
  3. Ministry of Land, Infrastructure and Transport (www.molit.go.kr)
  4. B. Kim, B. Choi, J. An, H. Lee, and E. Kim, "Prediction of Centerlane Violation for vehicle in opposite direction using Fuzzy Logic and Interacting Multiple Model," Journaol of The Korean Institute of Intelligent Systems, vol.23, no.5, pp.444-450, 2013. (In Korean) https://doi.org/10.5391/JKIIS.2013.23.5.444
  5. Ciuseppe De Nicolao, Antonella Ferrara, Luisa Giacomini, "Onboard sensor-based collision risk assessment to improve pedestrians' safety," IEEE Transactions on Vehicular Technology, Vol. 56, No. 5, 2007.
  6. H. Jung, S.-G. Kim, and Y.-G. Kim, "Use of Fuzzy technique for Calculating Degree of Collision Risk in Obstacle Avoidance of Unmanned Underwater Vehicles," Journaol of The Korean Institute of Intelligent Systems, vol.21, no.1, pp.112-119, 2011. (In Korean) https://doi.org/10.5391/JKIIS.2011.21.1.112
  7. Specht, Donald F., "A general regression neural networks," IEEE Transactions on Neural Networks, Vol. 2, No. 6, 1991.
  8. Alex J. Smola, Bernhard Scholkopf, "A tutorial on Support vector Regression," Statistics and computing, 2004.
  9. G.B. Huang, and Q.Y. Zhu, "Extreme learning machine: a new learning scheme of feedforward neural networks," In Proc. IEEE International Joint Conference on Neural Networks, vol. 2, pp. 985-990, 2004.
  10. B. Kim, S. Park, B. Choi, E. Kim, H. Lee, H. Kang,"Collision risk assessment for pedestrians' safety using neural network," Journal of Institute of Control, robotics and Systems, vol. 17, no. 1, pp. 6-11, 2011. (In Korean) https://doi.org/10.5302/J.ICROS.2011.17.1.6
  11. B. Kim, S. Park, B. Choi and E. Kim, "Intelligent collision risk assessment based on neural network ensemble," In Proc. of The Society of Instrument and Control Engineers Annual Conference (SICE2010), pp. 2893-2895, Taipei, Taiwan, August 2010
  12. B. Kim, B. Choi, E. Kim, C. Ko, J. Hwang, "Collision risk assessment algorithm using front-view vehicle information," In Proc. of Korean Society of Automotive Engineers, pp. 925-927, 2013. (In Korean)
  13. S. Park, B. Kim, E. Kim, H. Lee, and H. J. Kang, "Recursive probabilistic approach to collision risk assessment for pedestrians' safety," Journaol of The Korean Institute of Intelligent Systems, vol.21, no.4, pp.475-480, 2011. (In Korean) https://doi.org/10.5391/JKIIS.2011.21.4.475
  14. PreScan, Tassinternational (www.tassinternational.com/prescan )

Cited by

  1. Real-Time Rear-End Collision-Warning System Using a Multilayer Perceptron Neural Network vol.17, pp.11, 2016, https://doi.org/10.1109/TITS.2016.2537878