Goal-based Trajectory Prediction for improved Cross-Dataset Generalization

TL;DR

This paper presents a goal-based trajectory prediction method using a heterogeneous GNN that improves cross-dataset generalization in autonomous driving scenarios, addressing the challenge of deploying models in unseen environments.

Contribution

Introduction of a novel GNN architecture with goal classification that enhances generalization across different datasets in trajectory prediction.

Findings

Improved performance in cross-dataset evaluation from Argoverse2 to NuScenes.

Effective goal classification enhances trajectory prediction accuracy.

Demonstrated better generalization compared to existing models.

Abstract

To achieve full autonomous driving, a good understanding of the surrounding environment is necessary. Especially predicting the future states of other traffic participants imposes a non-trivial challenge. Current SotA-models already show promising results when trained on real datasets (e.g. Argoverse2, NuScenes). Problems arise when these models are deployed to new/unseen areas. Typically, performance drops significantly, indicating that the models lack generalization. In this work, we introduce a new Graph Neural Network (GNN) that utilizes a heterogeneous graph consisting of traffic participants and vectorized road network. Latter, is used to classify goals, i.e. endpoints of the predicted trajectories, in a multi-staged approach, leading to a better generalization to unseen scenarios. We show the effectiveness of the goal selection process via cross-dataset evaluation, i.e. training on Argoverse2 and evaluating on NuScenes.