CAO-RONet: A Robust 4D Radar Odometry with Exploring More Information from Low-Quality Points

2025-03-03

AI Summary

To address the severe degradation of odometry performance under adverse weather conditions—caused by sparsity and high noise in 4D millimeter-wave radar point clouds—this paper proposes an end-to-end robust ego-motion estimation framework. Methodologically, it introduces three key innovations: (1) local point cloud completion to densify sparse measurements; (2) context-aware hierarchical feature matching, incorporating multi-scale geometric modeling and correlation balancing; and (3) a sliding-window coupled optimizer leveraging historical motion priors to enhance geometric consistency. These components jointly improve correspondence density and structural fidelity among low-quality radar points. Evaluated on the View-of-Delft dataset, our method achieves approximately 50% higher accuracy than state-of-the-art radar-based odometry approaches, with localization precision approaching that of LiDAR-based baselines. This work establishes a new paradigm for reliable autonomous navigation in challenging weather conditions.

Technology Category

• Intelligent Robots: State Estimation
• Computer Vision: Motion & Tracking
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Location- and context-aware Web and WoT applications and services
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Recently, 4D millimetre-wave radar exhibits more stable perception ability than LiDAR and camera under adverse conditions (e.g. rain and fog). However, low-quality radar points hinder its application, especially the odometry task that requires a dense and accurate matching. To fully explore the potential of 4D radar, we introduce a learning-based odometry framework, enabling robust ego-motion estimation from finite and uncertain geometry information. First, for sparse radar points, we propose a local completion to supplement missing structures and provide denser guideline for aligning two frames. Then, a context-aware association with a hierarchical structure flexibly matches points of different scales aided by feature similarity, and improves local matching consistency through correlation balancing. Finally, we present a window-based optimizer that uses historical priors to establish a coupling state estimation and correct errors of inter-frame matching. The superiority of our algorithm is confirmed on View-of-Delft dataset, achieving around a 50% performance improvement over previous approaches and delivering accuracy on par with LiDAR odometry. Our code will be available.

Problem

• Enhance 4D radar odometry accuracy using low-quality points.
• Develop a learning-based framework for robust ego-motion estimation.
• Improve matching consistency and error correction in radar odometry.

Innovation

• Local completion for sparse radar points
• Context-aware hierarchical point matching
• Window-based optimizer with historical priors