A Multi-Graph Convolutional Approach for Short-Term Turning Movement Prediction at Intersections

Abstract: Traffic flow forecasting is a crucial first step in intelligent and proactive traffic management. Traffic flow parameters are volatile and uncertain which makes traffic flow forecasting a hard task to perform if the right forecasting model is not used. Additionally, the non-Euclidian data structure of traffic flow parameters is challenging to analyze both from a spatial and temporal perspective. State of art deep learning approaches adopts pure convolution, recurrent neural network, and hybrid method to achieve this objective efficiently. However, many of the approaches in the state-of-the-art literature rely on complex architectures that could make them hard to train. This in addition to the fact that such complex architecture could add to the black box nature of deep learning. This study introduces a novel deep learning architecture, referred to as multigraph convolution neural network (MGCNN), for corridor-wide turning movement prediction at intersections. The proposed architecture combines a multigraph structure, built to model temporal variations in traffic data, with a spectral convolution operation to support modeling the spatial variations in traffic data over the graphs. The proposed model was tested using a twenty-day flow and traffic control data collected from an arterial in downtown Chattanooga, TN, with 10 signalized intersections. The model’s ability to perform short-term predictions over 1, 2, 3, 4, and 5 minutes into the future was evaluated against four baseline state-of-the-art models. The results showed that our proposed model is superior to the other baseline models in predicting turning movements with an MSE of 0.9.