FASTopoWM: Fast-Slow Lane Segment Topology Reasoning with Latent World Models

TL;DR

FASTopoWM is a novel framework that enhances lane topology reasoning in autonomous driving by leveraging fast-slow systems and latent world models, significantly improving detection and perception accuracy over state-of-the-art methods.

Contribution

It introduces a unified fast-slow lane reasoning framework with latent world models, enabling parallel supervision and better temporal perception in autonomous driving.

Findings

Outperforms state-of-the-art in lane segment detection (37.4% vs. 33.6% mAP)

Improves centerline perception accuracy (46.3% vs. 41.5% OLS)

Enhances temporal perception robustness in lane topology reasoning.

Abstract

Lane segment topology reasoning provides comprehensive bird's-eye view (BEV) road scene understanding, which can serve as a key perception module in planning-oriented end-to-end autonomous driving systems. Existing lane topology reasoning methods often fall short in effectively leveraging temporal information to enhance detection and reasoning performance. Recently, stream-based temporal propagation method has demonstrated promising results by incorporating temporal cues at both the query and BEV levels. However, it remains limited by over-reliance on historical queries, vulnerability to pose estimation failures, and insufficient temporal propagation. To overcome these limitations, we propose FASTopoWM, a novel fast-slow lane segment topology reasoning framework augmented with latent world models. To reduce the impact of pose estimation failures, this unified framework enables parallel supervision of both historical and newly initialized queries, facilitating mutual reinforcement between the fast and slow systems. Furthermore, we introduce latent query and BEV world models conditioned on the action latent to propagate the state representations from past observations to the current timestep. This design substantially improves the performance of temporal perception within the slow pipeline. Extensive experiments on the OpenLane-V2 benchmark demonstrate that FASTopoWM outperforms state-of-the-art methods in both lane segment detection (37.4% v.s. 33.6% on mAP) and centerline perception (46.3% v.s. 41.5% on OLS).