🤖 AI Summary
Current graph neural networks (GNNs) used in autonomous driving trajectory prediction lack systematic evaluation and design guidance regarding their ability to model spatial interactions and temporal dynamics across different layers. This work systematically evaluates 19 GNN layers within a unified framework, integrating multi-head attention mechanisms with various aggregation strategies. The study reveals that sum aggregation consistently outperforms mean aggregation, and that incorporating distance-aware edge weighting alongside multi-head attention significantly enhances modeling capacity. Through extensive experiments, five superior layer combinations are identified, with ARMA, Chebyshev, and topology-aware layers consistently achieving state-of-the-art performance and substantially improving prediction accuracy. These findings lead to practical design principles for effective GNN architectures in trajectory forecasting.
📝 Abstract
Autonomous driving systems rely on precise trajectory prediction to plan safe and efficient movement. Graph Neural Networks (GNNs) have become a promising approach for modelling spatiotemporal interactions among road agents. However, designing GNN architectures for trajectory prediction remains non-standardized, with little guidance on which graph layers effectively capture spatial interactions and temporal dynamics. This paper offers a detailed comparative study of 19 graph layer types, focusing on their spatial and temporal processing capabilities to discover the most effective architectures for trajectory prediction. Within the explored hyperparameter setting, we highlight five standout layer combinations, with ARMA, Chebyshev, and topology-aware layers consistently performing better than others. Beyond performance metrics, our findings yield practical design principles: sum-based aggregation is more effective than mean-based methods, multi-head attention mechanisms enable richer interactions, and assigning different weights to different hop distances significantly improves prediction accuracy. These findings offer useful guidance for designing more interpretable and effective trajectory prediction models.