Vectorized Scenario Description and Motion Prediction for Scenario-Based Testing
This work addresses scenario-based testing for automated vehicles, offering an incremental improvement by enabling data merging and comparison across diverse scenarios.
The paper tackles the problem of testing automated vehicles by proposing a vectorized scenario description based on road geometry and vehicle trajectories, and shows that training the VectorNet model on merged data from three scenarios partially achieves lower prediction errors than regression models on separate data.
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.