Graph-Based Spatial-Temporal Convolutional Network for Vehicle Trajectory Prediction in Autonomous Driving
This work addresses the critical problem of accurate and efficient trajectory forecasting for autonomous vehicles, representing an incremental improvement over existing methods.
The paper tackles vehicle trajectory prediction for autonomous driving by proposing a graph-based spatial-temporal convolutional network (GSTCN) that integrates graph convolutional networks and CNNs to model interactions and temporal features, achieving state-of-the-art performance on real-world freeway datasets with improved prediction errors, model sizes, and inference speeds.
Forecasting the trajectories of neighbor vehicles is a crucial step for decision making and motion planning of autonomous vehicles. This paper proposes a graph-based spatial-temporal convolutional network (GSTCN) to predict future trajectory distributions of all neighbor vehicles using past trajectories. This network tackles the spatial interactions using a graph convolutional network (GCN), and captures the temporal features with a convolutional neural network (CNN). The spatial-temporal features are encoded and decoded by a gated recurrent unit (GRU) network to generate future trajectory distributions. Besides, we propose a weighted adjacency matrix to describe the intensities of mutual influence between vehicles, and the ablation study demonstrates the effectiveness of our proposed scheme. Our network is evaluated on two real-world freeway trajectory datasets: I-80 and US-101 in the Next Generation Simulation (NGSIM).Comparisons in three aspects, including prediction errors, model sizes, and inference speeds, show that our network can achieve state-of-the-art performance.