Transferable and Adaptable Driving Behavior Prediction
Abstract: While autonomous vehicles still struggle to solve challenging situations during on-road driving, humans have long mastered the essence of driving with efficient, transferable, and adaptable driving capability. The obvious gap between humans and autonomous vehicles keeps us wondering about the essence of how human learns to drive. Inspired by humans' cognition model and semantic understanding during driving in a hierarchical learning procedure, we propose HATN, a hierarchical framework to generate high-quality, transferable, and adaptable predictions for driving behaviors in multi-agent dense-traffic environments. Our hierarchical method consists of a high-level intention identification policy and a low-level trajectory generation policy. We introduce a novel semantic definition for the two policies and generic state representation for each policy, so that the hierarchical framework is transferable across different driving scenarios. Besides, our model is able to capture variations of driving behaviors among individuals and scenarios by an online adaptation module. We demonstrate our algorithms in the task of trajectory prediction for real traffic data at intersections and roundabouts from the INTERACTION dataset. Through extensive numerical studies, it is evident that our method significantly outperformed other methods in terms of prediction accuracy, transferability, and adaptability. Pushing the performance by a considerable margin, we also provide a cognitive view of understanding the driving behavior behind such improvement. We highlight that in the future, more research attention and effort are deserved for the transferability and adaptability of autonomous driving planning and prediction algorithms. It is not only due to the promising performance elevation, but more fundamentally, they are crucial for the scalable and general deployment of autonomous vehicles.
Submission Length: Long submission (more than 12 pages of main content)
Previous TMLR Submission Url: https://openreview.net/forum?id=OdpxxnH2Th¬eId=VYXuhDI6wN
Changes Since Last Submission: The previous submission reveals the authors' identity. In this submission, those contents has been removed.
Assigned Action Editor: ~Matthew_Walter1
Submission Number: 337
Loading