Distribution-aware Goal Prediction and Conformant Model-based Planning for Safe Autonomous Driving
This addresses the problem of safe autonomous driving in novel scenarios for researchers and developers, representing a strong incremental improvement with a novel method for a known bottleneck.
The paper tackles the limited generalizability of imitation learning in autonomous driving by proposing a modular architecture that factorizes the task into obstacle-aware perception, distribution-aware goal prediction, and model-based planning, achieving state-of-the-art results on the CARNOVEL benchmark in the CARLA simulator.
The feasibility of collecting a large amount of expert demonstrations has inspired growing research interests in learning-to-drive settings, where models learn by imitating the driving behaviour from experts. However, exclusively relying on imitation can limit agents' generalisability to novel scenarios that are outside the support of the training data. In this paper, we address this challenge by factorising the driving task, based on the intuition that modular architectures are more generalisable and more robust to changes in the environment compared to monolithic, end-to-end frameworks. Specifically, we draw inspiration from the trajectory forecasting community and reformulate the learning-to-drive task as obstacle-aware perception and grounding, distribution-aware goal prediction, and model-based planning. Firstly, we train the obstacle-aware perception module to extract salient representation of the visual context. Then, we learn a multi-modal goal distribution by performing conditional density-estimation using normalising flow. Finally, we ground candidate trajectory predictions road geometry, and plan the actions based on on vehicle dynamics. Under the CARLA simulator, we report state-of-the-art results on the CARNOVEL benchmark.