AMap: Distilling Future Priors for Ahead-Aware Online HD Map Construction

TL;DR

AMap introduces a novel ahead-aware mapping framework that distills future context into a lightweight model, significantly improving forward-region perception in online HD map construction for autonomous driving without increasing inference costs.

Contribution

The paper proposes a pioneering distillation approach that transfers future-aware knowledge into a current-frame model, addressing safety gaps in existing temporal fusion methods.

Findings

Outperforms state-of-the-art temporal models in forward regions

Enhances current-frame perception accuracy

Maintains efficiency comparable to single-frame inference

Abstract

Online High-Definition (HD) map construction is pivotal for autonomous driving. While recent approaches leverage historical temporal fusion to improve performance, we identify a critical safety flaw in this paradigm: it is inherently ``spatially backward-looking." These methods predominantly enhance map reconstruction in traversed areas, offering minimal improvement for the unseen road ahead. Crucially, our analysis of downstream planning tasks reveals a severe asymmetry: while rearward perception errors are often tolerable, inaccuracies in the forward region directly precipitate hazardous driving maneuvers. To bridge this safety gap, we propose AMap, a novel framework for Ahead-aware online HD Mapping. We pioneer a ``distill-from-future" paradigm, where a teacher model with privileged access to future temporal contexts guides a lightweight student model restricted to the current frame. This process implicitly compresses prospective knowledge into the student model, endowing it with ``look-ahead" capabilities at zero inference-time cost. Technically, we introduce a Multi-Level BEV Distillation strategy with spatial masking and an Asymmetric Query Adaptation module to effectively transfer future-aware representations to the student's static queries. Extensive experiments on the nuScenes and Argoverse 2 benchmark demonstrate that AMap significantly enhances current-frame perception. Most notably, it outperforms state-of-the-art temporal models in critical forward regions while maintaining the efficiency of single current frame inference.