Irregular Traffic Time Series Forecasting Based on Asynchronous Spatio-Temporal Graph Convolutional Network

TL;DR

This paper introduces ASeer, a novel neural network architecture designed to accurately forecast irregular traffic time series by capturing asynchronous spatial dependencies and irregular temporal patterns, outperforming existing methods.

Contribution

The paper presents a comprehensive framework combining asynchronous graph diffusion, personalized time encoding, time-aware convolution, and semi-autoregressive prediction for irregular traffic forecasting.

Findings

ASeer outperforms twelve baselines across six metrics.

The model effectively captures asynchronous spatial dependencies.

The approach handles variable-length sequence prediction efficiently.

Abstract

Accurate traffic forecasting is crucial for the development of Intelligent Transportation Systems (ITS), playing a pivotal role in modern urban traffic management. Traditional forecasting methods, however, struggle with the irregular traffic time series resulting from adaptive traffic signal controls, presenting challenges in asynchronous spatial dependency, irregular temporal dependency, and predicting variable-length sequences. To this end, we propose an Asynchronous Spatio-tEmporal graph convolutional nEtwoRk (ASeer) tailored for irregular traffic time series forecasting. Specifically, we first propose an Asynchronous Graph Diffusion Network to capture the spatial dependency between asynchronously measured traffic states regulated by adaptive traffic signals. After that, to capture the temporal dependency within irregular traffic state sequences, a personalized time encoding is devised to embed the continuous time signals. Then, we propose a Transformable Time-aware Convolution Network, which adapts meta-filters for time-aware convolution on the sequences with inconsistent temporal flow. Additionally, a Semi-Autoregressive Prediction Network, comprising a state evolution unit and a semi-autoregressive predictor, is designed to predict variable-length traffic sequences effectively and efficiently. Extensive experiments on a newly established benchmark demonstrate the superiority of ASeer compared with twelve competitive baselines across six metrics.