RadLoc: Radar-based 3-DoF Global Localization via Fast, Robust, and Lightweight Spatial Descriptor Across Diverse Environmental Scenarios

2026-07-09

AI Summary

This work addresses the challenge of balancing efficiency, robustness, and model compactness in radar-based global localization under adverse weather and complex environmental conditions. We propose an end-to-end 3-degree-of-freedom global localization framework that integrates 1D CA-CFAR filtering for accelerated preprocessing, leverages the radar’s near-range dominance to construct a compact spatial descriptor, and employs a hierarchical coarse-to-fine retrieval strategy. Pose estimation is efficiently achieved through phase correlation. Extensive experiments across 15 sequences from five diverse datasets demonstrate that our method achieves state-of-the-art robustness while maintaining the smallest descriptor size and fastest retrieval speed, significantly outperforming existing approaches.

Abstract

While global localization using spinning radar has gained attention for its robustness to adverse weather and challenging environments, many studies have focused on individual components such as place recognition or pose estimation. In this paper, we take a holistic view of radar sensor-based global localization and present RadLoc, a fast, robust, and lightweight end-to-end pipeline from place recognition to 3-DoF pose estimation. RadLoc accelerates pre-processing using 1D CA-CFAR filtering and leverages the near-range dominance in spinning radar images to design a compact descriptor and an efficient hierarchical coarse-to-fine retrieval strategy. Moreover, coupled with phase correlation-based 3-DoF pose estimation, it forms a versatile global localization module applicable to SLAM and multi-session SLAM systems. Extensive experiments on 15 sequences across 5 datasets demonstrate that RadLoc achieves robust performance while maintaining the smallest descriptor size and fastest retrieval time among state-of-the-art approaches. The supplementary materials are available at https://sparolab.github.io/research/radloc/.

Problem

• radar-based localization
• global localization
• 3-DoF pose estimation
• spinning radar
• environmental robustness

Innovation

• radar-based localization
• spatial descriptor
• 3-DoF pose estimation
• coarse-to-fine retrieval
• CA-CFAR filtering