Reducing hysteresis and over‐smoothing in traffic estimation: A multistream spatial‐temporal graph convolutional network

Abstract

Accurate traffic estimation contributes to safer route planning for Autonomous Vehicles (AVs). Currently, deep learning methods based on graph convolution networks (GCNs) and recurrent neural networks (RNNs) are widely used in traffic estimation. However, such methods suffer from spatial over‐smoothing and temporal hysteresis, which lead to estimation results deviating from the ground truth. Therefore, a multistream spatial‐temporal graph convolutional network (MSGCN) is proposed in this paper to deal with these issues. It integrates local, global and differential spatial‐temporal features which are modeled from multiple dimensions to deal with spatially correlation and time‐varying evolution of traffic states. Experimental results obtained on a real‐world dataset demonstrate the effectiveness of the proposed MSGCN. Furthermore, we measure the performance in both usual and unusual traffic states. Compared to baseline models, MSGCN is more accurate and robust in the two states.

Safety and security of autonomous vehicles

“…In the paper by Yu et al, 3 common deep learning methods used in traffic estimation suffer from spatial over-smoothing and temporal hysteresis, which lead to estimation results deviating from the ground truth. Therefore, the authors propose a multi-stream spatial-temporal graph convolutional network (MSGCN), integrating local, global, and differential spatial-temporal features that are modeled from multiple dimensions.…”