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. By mimicking humans' cognition model and semantic understanding during driving, 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 sub-task definition and generic state representation for each sub-task. With these techniques, 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 state-of-the-art 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 transferability and adaptability. It is not only due to the promising performance elevation of prediction and planning algorithms, but more fundamentally, they are crucial for the scalable and general deployment of autonomous vehicles.

Safety-Enhanced Autonomous Driving Using Interpretable Sensor Fusion Transformer

“…Note that while more advanced trajectory prediction methods [40,41,42] and safety controller [30,43] can be used, we empirically found our dynamics propagation with moving average and linear programming controller sufficient. In case of more complicated driving tasks, those advanced algorithms can be easily integrated into our framework in the future.…”

SHAIL: Safety-Aware Hierarchical Adversarial Imitation Learning for Autonomous Driving in Urban Environments

“…Much of recent work that performs vehicle behavior prediction hierarchically (e.g. [26]) can be easily extended to off-policy hierarchical imitation learning, as many policies learned to predict vehicle behavior when conditioned on a goal could be extended to control an ego vehicle. A flavor of on-policy hierarchical imitation learning has been applied to driving policies, in which long-horizon planning learned to mimics a query-able expert is interleaved with fast, short-horizon, low-level optimal control [27].…”

Efficient Game-Theoretic Planning with Prediction Heuristic for Socially-Compliant Autonomous Driving

“…Due to the fact that the possible interaction and driving patterns of the two vehicles are usually limited, the computation resources are mostly spared in uninformed search space, leading to low efficiency. On the other hand, nowadays many efficient learning-based prediction methods have been developed [10]- [12]. With actual driver interaction captured in real human data, the predicted trajectory from these learned prediction models usually presents feasible interaction patterns of the two agents, like one yielding car and one passing car.…”

“…We resort to the easily accessed and well established behavior prediction models [10]- [13], which can efficiently predict possible future behavior patterns. These behavior pattern predictions, though sometimes coarse or not highly accurate, can serve as an important heuristic to guide the search of MCTS, so that computational resources can be efficiently allocated at the valuable branches.…”