지능형 교통 시스템을 위한 Graph Neural Networks 기반 교통 속도 예측
Traffic Speed Prediction Based on Graph Neural Networks for Intelligent Transportation System
0
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
최근 활발히 연구되는 딥러닝 방법론은 인공지능의 성능을 급속도로 향상시켰고, 이에 따라다양한 산업 분야에서 딥러닝을 활용한 시스템이 제시되고 있다. 교통 시스템에서는 GNN을활용한 공간-시간 그래프 모델링이 교통 속도 예측에 효과적인 것으로 밝혀졌지만, 이는 메모리 병목 현상을 유발하기 때문에 모델이 비효율적으로 학습된다는 단점이 있다. 따라서 본 연구에서는 그래프 분할 방법을 통해 도로 네트워크를 분할하여 메모리 병목 현상을 완화함과동시에 우수한 성능을 달성하고자 한다. 제안 방법론을 검증하기 위해 인천시 UTIC 데이터분석 결과를 바탕으로 Jensen-Shannon divergence를 사용하여 도로 속도 분포의 유사도를 측정하였다. 그리고 측정된 유사도를 바탕으로 스펙트럴 클러스터링을 수행하여 도로 네트워크를군집화하였다. 성능 측정 결과, 도로 네트워크가 7개의 네트워크로 분할되었을 때 MAE 기준5.52km/h의 오차로 비교 모델 대비 가장 우수한 정확도를 보임과 동시에 메모리 병목 현상 또한 완화되는 것을 확인할 수 있었다.
Abstract
Deep learning methodology, which has been actively studied in recent years, has improved the performance of artificial intelligence. Accordingly, systems utilizing deep learning have been proposed in various industries. In traffic systems, spatio-temporal graph modeling using GNN was found to be effective in predicting traffic speed. Still, it has a disadvantage that the model is trained inefficiently due to the memory bottleneck. Therefore, in this study, the road network is clustered through the graph clustering algorithm to reduce memory bottlenecks and simultaneously achieve superior performance. In order to verify the proposed method, the similarity of road speed distribution was measured using Jensen-Shannon divergence based on the analysis result of Incheon UTIC data. Then, the road network was clustered by spectrum clustering based on the measured similarity. As a result of the experiments, it was found that when the road network was divided into seven networks, the memory bottleneck was alleviated while recording the best performance compared to the baselines with MAE of 5.52km/h.
참고문헌 (22)
- Endres, "A New Metric for Probability Distributions", IEEE Transactions on Information Theory, 49(7), pp.1858~1860 (2003)
- Fu, R., "Using LSTM and GRU Neural Network Methods for Traffic Flow Prediction", Youth Academic Annual Conference of Chinese Association of Automation, pp.324~328 (2016)
- George, K., "A Fast and High Quality Multilevel Scheme for Partitioning Irregular Graphs", SIAM Journal on scientific Computing, 20(1), pp.359~392 (1998)
- Hinton, G. E., "A Fast Learning Algorithm for Deep Belief Nets", Neural Computation, 18(7), pp.1527~1554 (2006)
- 김동현, "심층인공신경망(DNN)과 다각도 상황 정보 기반의 서울시 도로 링크별 교통 혼잡도 예측", 한국ITS학회 논문지, 18(4), pp.44~57 (2019), 한국ITS학회[DOI]
- 김영찬, "시공간적 영향력을 반영한 딥러닝 기반의 통행속도 예측 모형 개발", 한국ITS학회 논문지, 19(1), pp.1~16 (2020), 한국ITS학회[DOI]
- "K-indicator" (2020)[URL]
- Kipf, T. N., "Semi-Supervised Classification with Graph Convolutional Networks", arXiv preprint arXiv:1609.02907, ~ (2016)
- LeCun, Y., "Deep Learning", Nature, 521(7553), pp.436~444 (2015)
- Li, Y., "Diffusion Convolutional Recurrent Neural Network:Data-Driven Traffic Forecasting", The International Conference on Learning Representations (2018)
- Lv, Z., "Lc-rnn: A Deep Learning Model for Traffic Speed Prediction", International Joint Conferences on Artificial Intelligence, pp.3470~3476 (2018)
- Ma, X., "Learning Traffic as Images: A Deep Convolutional Neural Network for Large-Scale Transportation Network Speed Prediction", Sensors, 4(17), pp.818~ (2017)
- Mallick, T., "Graph-Partitioning-Based Diffusion Convolution Recurrent Neural Network for Large-Scale Traffic Forecasting", Transportation Research Record, 2674(9), pp.473~488 (2020)
- Oord, A. Dieleman, S., "Wavenet: A Generative Model for Raw Audio", arXiv preprint arXiv:1609.03499, ~ (2016)
- C. Park, "STGRAT: A Spatio-Temporal Graph Attention Network for Traffic Forecasting", arXiv preprint arXiv:1911.13181, ~ (2019)
- Wang, J., "Traffic Speed Prediction and Congestion Source Exploration: A Deep Learning Method", IEEE 16th International Conference on Data Mining, pp.499~508 (2016)
- Wu, Z., "Graph Wavenet for Deep Spatial-Temporal Graph Modeling", International Joint Conference on Artificial Intelligence (2019)
- Lv, Y., "Traffic Flow Prediction with Big Data:A Deep Learning Approach", IEEE Transactions on Intelligent Transportation Systems, 2(16), pp.865~873 (2014)
- Yu, B., "Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecasting", International Joint Conference on Artificial Intelligence, pp.3634~3640 (2018)
- Yu, H., "Spatiotemporal Recurrent Convolutional Networks for Traffic Prediction in Transportation Networks", Sensors, 7(17), pp.1501~ (2017)
- Zhang, J., "Gaan: Gated Attention Networks for Learning on Large and Spatiotemporal Graphs" (2018), Uncertainty in Artificial Intelligence
- Zhao, Z., "LSTM Network: A Deep Learning Approach for Short-Term Traffic Forecast", IET Intelligent Transport Systems, 2(11), pp.68~75 (2017)