Dr-PoGO: Direct Radar Pose-Graph Optimization
Cedric Le Gentil, Weican Li, Leonardo Brizi, Timothy D. Barfoot
TLDR
Dr-PoGO is a novel radar-based SLAM method using direct registration and pose-graph optimization for robust localization in all weather conditions.
Key contributions
- Leverages direct registration for radar odometry and loop-closure, avoiding feature extraction.
- Integrates RaPlace for place recognition, with a coarse-to-fine registration for initial loop-closure transforms.
- Optimizes the global trajectory using a robust pose-graph optimization framework.
- Achieves state-of-the-art radar SLAM performance over 300km in diverse automotive environments.
Why it matters
This paper introduces a robust radar SLAM solution, Dr-PoGO, that excels in adverse weather where cameras and lidars fail. By using direct registration and pose-graph optimization, it significantly improves localization accuracy for autonomous systems in challenging real-world conditions.
Original Abstract
This paper introduces Dr-PoGO, a method for Simultaneous Localization And Mapping (SLAM) using a 2D spinning radar. Unlike cameras or lidars that require line-of-sight, millimetre-wave radars can `see' through dust, falling snow, rain, etc. Accordingly, it is a great modality for robust perception regardless of the weather conditions. While most existing radar-based SLAM methods rely on the extraction of point clouds or features to perform ego-motion estimation, Dr-PoGO leverages direct registration techniques for odometry (DRO) and loop-closure registration. An off-the-shelf radar-focused place recognition algorithm, RaPlace, provides loop-closure candidates. As RaPlace does not provide relative transformations, Dr-PoGO introduces a coarse-to-fine registration that uses visual features and descriptors to obtain an initial guess for the direct transformation refinement. The global trajectory is optimized in a pose-graph optimization. Dr-PoGO demonstrates state-of-the-art performance over 300km of data in various real-world automotive environments. Our implementation is publicly available: https://github.com/utiasASRL/dr_pogo.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.