NEAT: Neural Attention Fields for End-to-End Autonomous Driving
This addresses the challenge of interpretable and robust end-to-end driving models for autonomous vehicles, with incremental improvements in performance and explainability.
The paper tackles the problem of efficient scene reasoning for autonomous driving by introducing NEural ATtention fields (NEAT), a continuous representation that maps locations to waypoints and semantics, achieving driving scores on par with a privileged expert in adverse conditions.
Efficient reasoning about the semantic, spatial, and temporal structure of a scene is a crucial prerequisite for autonomous driving. We present NEural ATtention fields (NEAT), a novel representation that enables such reasoning for end-to-end imitation learning models. NEAT is a continuous function which maps locations in Bird's Eye View (BEV) scene coordinates to waypoints and semantics, using intermediate attention maps to iteratively compress high-dimensional 2D image features into a compact representation. This allows our model to selectively attend to relevant regions in the input while ignoring information irrelevant to the driving task, effectively associating the images with the BEV representation. In a new evaluation setting involving adverse environmental conditions and challenging scenarios, NEAT outperforms several strong baselines and achieves driving scores on par with the privileged CARLA expert used to generate its training data. Furthermore, visualizing the attention maps for models with NEAT intermediate representations provides improved interpretability.