http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
그래프 합성곱을 활용한 교통량 기반 미래 교통 속도 예측
김남혁(Namhyuk Kim),허자윤(Jayun Huh),김태헌(Taeheon Kim),박성환(Sunghwan Park) 한국자동차공학회 2023 한국자동차공학회 부문종합 학술대회 Vol.2023 No.5
Navigation, one of the automotive electronic systems, provides the users with the optimal route to reach their destination and gives an estimated arrival time. In order to provide the optimal route, all possible routes to the destination within the constraints should be evaluated, which is calculated with predictive speed of the road network. Therefore, the reliable and accurate speed prediction of the road network is a prerequisite for delivering the optimal route to users. In this paper, we propose a novel road network speed prediction framework consisting of a traffic flow prediction model based on a Graph Convolutional Networks (GCN) and a travel time prediction model based on a naïve Convolutional Networks (CNN). First, the regional road network is formed as a graph data structure, and the graph data becomes input data of the traffic flow prediction model with traffic volume data. The output of the model is converted into travel time through the second model, the travel time prediction model. To evaluate the performance of the model, the model learned the actual traffic data from Seoul, and the output of the model was compared to the result from a baseline model.