Vectorized Scenario Description and Motion Prediction for Scenario-Based Testing

TL;DR

This paper introduces a vectorized scenario description method for automated vehicle testing, enabling unified representation of diverse scenarios and improving motion prediction accuracy through a novel model called VectorNet.

Contribution

The paper proposes a new vectorized scenario description that merges diverse scenario data and enhances motion prediction for AVs, surpassing traditional regression models.

Findings

VectorNet predicts AV trajectories with lower errors than regression models.

The vectorized description captures spatial and temporal nuances effectively.

The method allows comparison and combination of test and real-world scenarios.

Abstract

Automated vehicles (AVs) are tested in diverse scenarios, typically specified by parameters such as velocities, distances, or curve radii. To describe scenarios uniformly independent of such parameters, this paper proposes a vectorized scenario description defined by the road geometry and vehicles' trajectories. Data of this form are generated for three scenarios, merged, and used to train the motion prediction model VectorNet, allowing to predict an AV's trajectory for unseen scenarios. Predicting scenario evaluation metrics, VectorNet partially achieves lower errors than regression models that separately process the three scenarios' data. However, for comprehensive generalization, sufficient variance in the training data must be ensured. Thus, contrary to existing methods, our proposed method can merge diverse scenarios' data and exploit spatial and temporal nuances in the vectorized scenario description. As a result, data from specified test scenarios and real-world scenarios can be compared and combined for (predictive) analyses and scenario selection.