Map-Adaptive Goal-Based Trajectory Prediction
This improves autonomous driving systems by enabling more accurate long-term predictions that generalize better to new road scenes.
The paper tackles vehicle trajectory prediction by generating goal paths from lane centerlines and predicting trajectories along these paths, achieving state-of-the-art performance on a 6-second horizon across internal and nuScenes datasets.
We present a new method for multi-modal, long-term vehicle trajectory prediction. Our approach relies on using lane centerlines captured in rich maps of the environment to generate a set of proposed goal paths for each vehicle. Using these paths -- which are generated at run time and therefore dynamically adapt to the scene -- as spatial anchors, we predict a set of goal-based trajectories along with a categorical distribution over the goals. This approach allows us to directly model the goal-directed behavior of traffic actors, which unlocks the potential for more accurate long-term prediction. Our experimental results on both a large-scale internal driving dataset and on the public nuScenes dataset show that our model outperforms state-of-the-art approaches for vehicle trajectory prediction over a 6-second horizon. We also empirically demonstrate that our model is better able to generalize to road scenes from a completely new city than existing methods.