Multi-Head Attention based Probabilistic Vehicle Trajectory Prediction
This addresses trajectory prediction for autonomous vehicles, but appears incremental as it builds on existing attention-based methods.
The paper tackles vehicle trajectory prediction by proposing a multi-head attention encoder-decoder model that generates probabilistic trajectories for multiple vehicles in parallel, showing clear improvements in positional error on both longitudinal and lateral directions in highway experiments.
This paper presents online-capable deep learning model for probabilistic vehicle trajectory prediction. We propose a simple encoder-decoder architecture based on multi-head attention. The proposed model generates the distribution of the predicted trajectories for multiple vehicles in parallel. Our approach to model the interactions can learn to attend to a few influential vehicles in an unsupervised manner, which can improve the interpretability of the network. The experiments using naturalistic trajectories at highway show the clear improvement in terms of positional error on both longitudinal and lateral direction.