Exploring Contextual Representation and Multi-Modality for End-to-End Autonomous Driving

TL;DR

This paper introduces a multi-modal, context-aware end-to-end autonomous driving framework inspired by human neural maps, integrating sensor fusion and transformer-based encoding to improve hazard anticipation and decision-making.

Contribution

It formalizes sensor fusion within an end-to-end system using a vision transformer for contextual understanding, achieving superior accuracy and driving performance.

Findings

Achieves 0.67m displacement error, 6.9% better than current methods on nuScenes.

Improves route completion and reduces infractions in CARLA benchmarks.

Enhances open and closed-loop autonomous driving performance.

Abstract

Learning contextual and spatial environmental representations enhances autonomous vehicle's hazard anticipation and decision-making in complex scenarios. Recent perception systems enhance spatial understanding with sensor fusion but often lack full environmental context. Humans, when driving, naturally employ neural maps that integrate various factors such as historical data, situational subtleties, and behavioral predictions of other road users to form a rich contextual understanding of their surroundings. This neural map-based comprehension is integral to making informed decisions on the road. In contrast, even with their significant advancements, autonomous systems have yet to fully harness this depth of human-like contextual understanding. Motivated by this, our work draws inspiration from human driving patterns and seeks to formalize the sensor fusion approach within an end-to-end autonomous driving framework. We introduce a framework that integrates three cameras (left, right, and center) to emulate the human field of view, coupled with top-down bird-eye-view semantic data to enhance contextual representation. The sensor data is fused and encoded using a self-attention mechanism, leading to an auto-regressive waypoint prediction module. We treat feature representation as a sequential problem, employing a vision transformer to distill the contextual interplay between sensor modalities. The efficacy of the proposed method is experimentally evaluated in both open and closed-loop settings. Our method achieves displacement error by 0.67m in open-loop settings, surpassing current methods by 6.9% on the nuScenes dataset. In closed-loop evaluations on CARLA's Town05 Long and Longest6 benchmarks, the proposed method enhances driving performance, route completion, and reduces infractions.