Exploring rounD Dataset for Domain Generalization in Autonomous Vehicle Trajectory Prediction

Zhang, Zikai

Abstract

This paper analyzes the rounD dataset to advance motion forecasting algorithms for autonomous vehicles navigating complex roundabout environments. We develop a trajectory prediction framework inspired by Gated Recurrent Unit (GRU) networks and graph-based modules to effectively model vehicle interactions. Our primary objective is to evaluate the generalizability of the proposed model across diverse training and testing datasets. Through extensive experiments, we investigate how varying data distributions—such as different road configurations and recording times—impact the model’s prediction accuracy and robustness. This study provides key insights into the challenges of domain generalization in autonomous vehicle trajectory prediction.

Citation

Zhang, Z. (2024). Exploring rounD Dataset for Domain Generalization in Autonomous Vehicle Trajectory Prediction. Sensors, 24(23), Article 7538. https://doi.org/10.3390/s24237538

• Journal Article Type: Article
• Acceptance Date: Nov 21, 2024
• Online Publication Date: Nov 26, 2024
• Publication Date: Nov 26, 2024
• Deposit Date: Dec 18, 2024
• Publicly Available Date: Dec 18, 2024
• Journal: Sensors
• Electronic ISSN: 1424-8220
• Publisher: MDPI
• Peer Reviewed: Peer Reviewed
• Volume: 24
• Issue: 23
• Article Number: 7538
• DOI: https://doi.org/10.3390/s24237538
• Keywords: motion forecasting, machine learning, domain generalization, driving behavior, trajectory prediction
• Public URL: https://durham-repository.worktribe.com/output/3221334