RLPR: Radar-to-LiDAR Place Recognition via Two-Stage Asymmetric Cross-Modal Alignment for Autonomous Driving

TL;DR

This work proposes RLPR, a robust radar-to-LiDAR place recognition framework compatible with single-chip, scanning, and 4D radars, and introduces a two-stage asymmetric cross-modal alignment (TACMA) strategy, which leverages the pre-trained radar branch as a discriminative anchor to guide the alignment process.

Abstract

All-weather autonomy is critical for autonomous driving, which necessitates reliable localization across diverse scenarios. While LiDAR place recognition is widely deployed for this task, its performance degrades in adverse weather. Conversely, radar-based methods, though weather-resilient, are hindered by the general unavailability of radar maps. To bridge this gap, radar-to-LiDAR place recognition, which localizes radar scans within existing LiDAR maps, has garnered increasing interest. However, extracting discriminative and generalizable features shared between modalities remains challenging, compounded by the scarcity of large-scale paired training data and the signal heterogeneity across radar types. In this work, we propose RLPR, a robust radar-to-LiDAR place recognition framework compatible with single-chip, scanning, and 4D radars. We first design a dual-stream network to extract structural features that abstract away from sensor-specific signal properties (e.g., Doppler or RCS). Subsequently, motivated by our task-specific asymmetry observation between radar and LiDAR, we introduce a two-stage asymmetric cross-modal alignment (TACMA) strategy, which leverages the pre-trained radar branch as a discriminative anchor to guide the alignment process. Experiments on four datasets demonstrate that RLPR achieves state-of-the-art recognition accuracy with strong zero-shot generalization capabilities.

Figures (6)

• Figure 1: Conceptual overview of the proposed RLPR framework. (a) RLPR facilitates robust place recognition across heterogeneous radar types. (b) Existing methods perform symmetric feature alignment. (c) Motivated by the task‑specific asymmetry between radar and LiDAR, we design a two-stage asymmetric cross-modal alignment (TACMA) strategy that enables effective feature alignment.
• Figure 2: The system overview of RLPR. The proposed radar-to-LiDAR place recognition framework consists of a dual-stream network and the two-stage asymmetric cross-modal alignment (TACMA) strategy. The dual-stream network is designed to extract global and local geometric features, while the TACMA strategy exploits cross-modal commonalities for accurate and generalizable retrieval.
• Figure 3: Architecture of the proposed Polar Context Enhancer. It captures global dependencies through two-way scanning along range and azimuth axes, generating an importance map to gate the polar BEV. The corresponding images share the same color scale.
• Figure 4: Conditional entropy analysis and feature map visualizations. A comparison of the feature maps shows that pre-training enables the extraction of distinct geometric saliency. Meanwhile, the intrinsic asymmetry between LiDAR and Radar drives a divergence in marginal entropy after pre-training, which leads to the observed reversal in conditional entropy. All visualizations are channel-averaged and share the same color scale.
• Figure 5: A comparative visualization of radar-to-radar and LiDAR-to-LiDAR place recognition accuracy.