RESBev: Making BEV Perception More Robust

TL;DR

RESBev is a robust, plug-and-play method that enhances bird's-eye-view perception in autonomous driving by predicting clean features from corrupted observations, improving safety under disturbances and attacks.

Contribution

It introduces a latent semantic prediction framework that boosts BEV perception robustness without altering existing models, applicable to natural and adversarial disturbances.

Findings

Significantly improves robustness against disturbances and attacks

Operates at the semantic feature level of BEV perception pipeline

Requires only few-shot fine-tuning for effective performance

Abstract

Bird's-eye-view (BEV) perception has emerged as a cornerstone of autonomous driving systems, providing a structured, ego-centric representation critical for downstream planning and control. However, real-world deployment faces challenges from sensor degradation and adversarial attacks, which can cause severe perceptual anomalies and ultimately compromise the safety of autonomous driving systems. To address this, we propose a resilient and plug-and-play BEV perception method, RESBev, which can be easily applied to existing BEV perception methods to enhance their robustness to diverse disturbances. Specifically, we reframe perception robustness as a latent semantic prediction problem. A latent world model is constructed to extract spatiotemporal correlations across sequential BEV observations, thereby learning the underlying BEV state transitions to predict clean BEV features for reconstructing corrupted observations. The proposed framework operates at the semantic feature level of the Lift-Splat-Shoot pipeline, enabling recovery that generalizes across both natural disturbances and adversarial attacks without modifying the underlying backbone. Extensive experiments on the nuScenes dataset demonstrate that, with few-shot fine-tuning, RESBev significantly improves the robustness of existing BEV perception models against various external disturbances and adversarial attacks.