EFEAR-4D: Ego-Velocity Filtering for Efficient and Accurate 4D Radar Odometry

2024-05-16

IEEE Robotics and Automation Letters

AI Summary

To address the degradation of conventional sensor-based odometry (e.g., camera, LiDAR, IMU) under adverse weather conditions such as snow and fog, this paper proposes a learning-free, robust 4D millimeter-wave (mmWave) radar odometry method. Methodologically, it introduces— for the first time—a Doppler-velocity-based dynamic motion prior estimation mechanism; designs a dynamic object removal strategy coupled with region-adaptive feature extraction to enhance matching robustness under sparse, noisy point clouds; and incorporates tightly-coupled IMU integration. Furthermore, it systematically analyzes the critical impact of radar mounting height on point cloud quality and localization accuracy. Evaluated on two public 4D radar datasets, the method achieves state-of-the-art accuracy and reliability, demonstrating both effectiveness under extreme environmental conditions and feasibility for real-world engineering deployment.

Technology Category

• Intelligent Robots: State Estimation
• Computer Vision: Motion & Tracking
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Odometry is a crucial component for successfully implementing autonomous navigation, relying on sensors such as cameras, LiDARs and IMUs. However, these sensors may encounter challenges in extreme weather conditions, such as snowfall and fog. The emergence of FMCW radar technology offers the potential for robust perception in adverse conditions. As the latest generation of FWCW radars, the 4D mmWave radar provides point cloud with range, azimuth, elevation, and Doppler velocity information, despite inherent sparsity and noises in the point cloud. EFEAR-4D exploits Doppler velocity information delicately for robust ego-velocity estimation, resulting in a highly accurate prior guess. EFEAR-4D maintains robustness against point-cloud sparsity and noises across diverse environments through dynamic object removal and effective region-wise feature extraction. Extensive experiments on two publicly available 4D radar datasets demonstrate state-of-the-art reliability and localization accuracy of EFEAR-4D under various conditions. Furthermore, we have collected a dataset following the same route but varying installation heights of the 4D radar, emphasizing the significant impact of radar height on point cloud quality, a crucial consideration for real-world deployments.

Problem

• Improves 4D radar odometry accuracy in extreme weather
• Addresses point cloud sparsity and noise challenges
• Evaluates radar height impact on point cloud quality

Innovation

• Uses Doppler velocity for ego-velocity estimation
• Dynamic object removal for robustness
• Region-wise feature extraction for accuracy