Vehicular Speed Prediction Method for Highway Scenarios Based on Spatiotemporal Graph Convolutional Networks and Potential Field Theory

Traffic flow analysis largely depends on accurate predictions of microscopic speed. Due to the complexity and stochastic of real-world driving environments, traditional model-driven methods face significant challenges. In recent years, data-driven methods that combine advanced intelligent algorithms have emerged to address the issue of vehicle speed prediction. However, most existing studies primarily focus on the static spatiotemporal relationships between vehicles, with only a few exploring dynamic spatiotemporal correlations. In this study, a new graph structure, named the vehicle generalized-dynamic graph, is constructed to characterize the vehicle trajectory, emphasizing the spatial and temporal evolution of vehicles. In the spatial dimension, we employ the potential field theory to compute the field strength of vehicles in different motion states, enhancing the description of interdependencies among vehicles. In the temporal dimension, the movements of each vehicle at different time points are interrelated. Furthermore, we propose a novel neural network architecture, named the dynamic edge graph convolutional network (DEGCN), to address the vehicle speed prediction problem. The DEGCN model is evaluated through experiments on three real-world vehicle trajectory datasets. The experimental results demonstrate that the proposed model outperforms other baseline models in terms of prediction performance. Additionally, various ablation studies are conducted to evaluate the effectiveness of different components, the potential field theory, and the vehicle generalized-dynamic graph structure.

I. Introduction

Nowadays with the fast progress of artificial intelligence (AI) technology and the increase of high-quality trajectory data, it becomes feasible to build an accurate model to predict the vehicle speed [1], [2], [3]. The core task of vehicle speed prediction is to mine and explore collected vehicle trajectory data, extract temporal and spatial correlation features between surrounding vehicles, and predict the vehicle lateral speed and longitudinal speed in the future [4], [5]. Moreover, the accurate prediction results can further help the autonomous vehicle to make reasonable decision planning and enhance traffic safety and driving comfort in a complex traffic environment [6], [7], [8].