NEAT: Neural Attention Fields for End-to-End Autonomous Driving

09/09/2021
by   Kashyap Chitta, et al.
20

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.

READ FULL TEXT

page 7

page 8

research
05/11/2022

NMR: Neural Manifold Representation for Autonomous Driving

Autonomous driving requires efficient reasoning about the Spatio-tempora...
research
09/14/2023

What Matters to Enhance Traffic Rule Compliance of Imitation Learning for Automated Driving

More research attention has recently been given to end-to-end autonomous...
research
07/10/2020

Learning Accurate and Human-Like Driving using Semantic Maps and Attention

This paper investigates how end-to-end driving models can be improved to...
research
11/21/2018

Early Fusion for Goal Directed Robotic Vision

Increasingly, perceptual systems are being codified as strict pipelines ...
research
10/25/2022

PlanT: Explainable Planning Transformers via Object-Level Representations

Planning an optimal route in a complex environment requires efficient re...
research
02/07/2023

Explainable Action Prediction through Self-Supervision on Scene Graphs

This work explores scene graphs as a distilled representation of high-le...
research
05/18/2022

CARNet: A Dynamic Autoencoder for Learning Latent Dynamics in Autonomous Driving Tasks

Autonomous driving has received a lot of attention in the automotive ind...

Please sign up or login with your details

Forgot password? Click here to reset