The Journal of The Korea Institute of Intelligent Transport Systems (한국ITS학회 논문지)

The Korea Institute of Intelligent Transport Systems (한국ITS학회)
권호 표지
DOI QR코드

DOI QR Code

Automated Vehicle Research by Recognizing Maneuvering Modes using LSTM Model

LSTM 모델 기반 주행 모드 인식을 통한 자율 주행에 관한 연구

  • Kim, Eunhui (CGS Graduate School of Green Transportation / Korea Advanced Institute of Science Technology(KAIST)) ;
  • Oh, Alice (School of Computing / Korea Advanced Institute of Science Technology(KAIST))
  • 김은희 (한국과학기술원 조천식교통대학원) ;
  • 오혜연 (한국과학기술원 전산학부)
  • Received : 2017.05.10
  • Accepted : 2017.07.24
  • Published : 2017.08.31
Download PDF
⟨ Previous Next ⟩

Abstract

This research is based on the previous research that personally preferred safe distance, rotating angle and speed are differentiated. Thus, we use machine learning model for recognizing maneuvering modes trained per personal or per similar driving pattern groups, and we evaluate automatic driving according to maneuvering modes. By utilizing driving knowledge, we subdivided 8 kinds of longitudinal modes and 4 kinds of lateral modes, and by combining the longitudinal and lateral modes, we build 21 kinds of maneuvering modes. we train the labeled data set per time stamp through RNN, LSTM and Bi-LSTM models by the trips of drivers, which are supervised deep learning models, and evaluate the maneuvering modes of automatic driving for the test data set. The evaluation dataset is aggregated of living trips of 3,000 populations by VTTI in USA for 3 years and we use 1500 trips of 22 people and training, validation and test dataset ratio is 80%, 10% and 10%, respectively. For recognizing longitudinal 8 kinds of maneuvering modes, RNN achieves better accuracy compared to LSTM, Bi-LSTM. However, Bi-LSTM improves the accuracy in recognizing 21 kinds of longitudinal and lateral maneuvering modes in comparison with RNN and LSTM as 1.54% and 0.47%, respectively.

본 연구에서는 운전자 별로 생활 중에 이동하는 주행 도로의 특징 및 교통상황이 서로 다르며 운전습관이 상이함을 고려하여, 운전자 혹은 운전자 그룹별 기계학습모형을 구성하고, 학습된 모델을 분석하여 운전자의 주행모드 별 특징을 탐색하여 자율 주행 자동차를 시뮬레이션 하였다. 운전지식을 활용하여 주행조작 전후 센서의 동작 상황에 따라 8종류의 종방향 모드와 4종류 회전모드로 구분하고, 종방향 모드와 회전모드를 결합한 21개의 결합형 주행모드로 세분화 하였다. 주행모드가 레이블 된 시계열 데이터에 대해 딥러닝 지도학습 모델인 RNN (Recurrent Neural Network), LSTM (Long Short-Term Memory), Bi-LSTM 모델을 활용하여서 운전자 별 혹은 운전자 그룹별 주행데이터를 학습하고, 학습된 모델을 테스트 데이터 셋에서 주행 모드인식률을 검증하였다. 실험 데이터는 미국 VTTI 기관에서 수집된 22명의 운전자의 1,500개의 실생활 주행 데이터가 사용되었다. 주행 모드 인식에 있어, 데이터 셋에 대해 Bi-LSTM 모델이 RNN, LSTM 모델에 비해 향상된 성능을 보였으며, 최대 93.41%의 주행모드 인식률을 확인하였다.

Keywords

References

  1. Alex G.(2012), "Supervised Sequence Labelling with Recurrent Neural Net-works," Chapter 3, 4, Springer.
  2. Choi J.(2015), "The role and expectation of the Vehicle ITS(Intelligent Transportation System)," Korea Evaluation Institute of Industrial Technology Issue Report.
  3. Gonzalez D., Perez J., Milanes V. et al.(2016), "A Review of Motion Planning Techniques for Automated Vehicles," IEEE transactions on intelligent transportation system, vol. 17, no. 4.
  4. Greff K., Srivastava R. K., Koutn J. et al.(2016), "LSTM: A Search Space Odyssey," IEEE Trs. on Nerual Networks and Learning Systems, no. 99.
  5. Hochreiter S. and Schmidhuber J.(1997), "Long short-term memory," Neural Computation, vol. 9, no. 8, pp.1735-1780. https://doi.org/10.1162/neco.1997.9.8.1735
  6. Jain A., Singh A., Koppula H. et al.(2016), "Recurrent neural networks for driver activity anticipation via sensory-fusion architecture," IEEE Intern. Conf. on Robotics and Automation (ICRA).
  7. Kim M., Moon J. and Lee J.(2015), "European automatic driving vehicle technology roadmap analysis (EopSS, ERTRAC)," Korea Evaluation Institute of Industrial Technology, Issue 3.
  8. Kuderer M., Gulati S. and Burgard W.(2015), "Learning driving styles for autonomous vehicles from demonstration," IEEE Inter. Conf. on Robotics and Automation (ICRA), pp.2641-2646.
  9. Lei Z., Cheng D., Liu Y., Qin D. Y. et al.(2017), "A Dynamic Control Strategy for Hybrid Electric Vehicles Based on Parameter Optimization for Multiple Driving Cycles and Driving Pattern Recognition," Energies, vol. 10, no. 1, p.54. https://doi.org/10.3390/en10010054
  10. Li G., Zhang H., Tang G. and Xie Y.(2014), "Maneuver modes analysis for hypersonic glide vehicles," IEEE Chinese Guidance, Navigation and Control Conf., pp.543-548.
  11. Ma X., Tao Z., Wang Y., Yu H. et al.(2015), "Long short-term memory neural network for traffic speed prediction using remote microwave sensor data," Transport. Res. Part C: Emerg. Technol., vol. 54, pp.187-197. https://doi.org/10.1016/j.trc.2015.03.014
  12. Takano W., Matsushita A., Iwao K. et al.(2008), "Recognition of human driving behaviors based on stochastic symbolization of time series signal," IEEE International Conf. on Intelligent Robots and Systems, pp.167-172.
  13. Tesla Third Quarter 2016 Update, http://files.shareholder.com/downloads/ABEA-4CW8X0/3100696065x0x913801/F9E5C36A-AFDD-4FF2-A375-ED9B0F912622/Q3_16_Update_Letter_-_final.pdf., 2017.04.21.
  14. Weninger F., Bergemann J. and Schulller B.(2015), "Introducing CURRENNT: The Munich Open-Source CUDA Recurrent Neural Network Toolkit," Journal of Machine Learning Research, vol. 15, pp.547-551.