Dr-lisa improves radar and lidar localization in complex 3d environments
Dr-LiSA: Direct Radar-Lidar Scan Alignment for $SE(3)$ Localization
Robotics
Summary
Localizing a vehicle or robot accurately is hard when combining different sensors like radar and lidar because they see the world very differently. The authors developed Dr-LiSA, a new way to directly compare radar readings to lidar maps by predicting what the radar should see from the lidar data. This method can locate the radar in full 3D space, going beyond previous methods that worked only on flat planes. Their approach works better than past radar-lidar methods and matches the accuracy of top radar-only systems on real driving data.
What this means in practice
- •For autonomous vehicle engineers: Improve vehicle localization accuracy in adverse weather by directly aligning radar scans with 3D lidar maps in full 3D space.
- •For robotics navigation teams: Use direct radar-lidar scan alignment to enhance 3D pose estimation for robots operating in mixed environmental conditions.
Authors
Alex Zhang, Daniil Lisus, Cedric Le Gentil, Timothy D. Barfoot
Abstract
This paper introduces Dr-LiSA, a first-of-its-kind direct method for localizing 2D spinning radar intensity measurements in $SE(3)$ against 3D lidar maps. Radar-lidar localization combines the complementary strengths of the two sensing modalities: radar is robust to adverse weather and precipitation, while lidar provides high-fidelity 3D maps in favourable conditions. However, existing radar-lidar localization methods are restricted to planar $SE(2)$ localization and have generally fallen short of the accuracy achieved by lidar-lidar and even radar-radar systems. A key challenge is the substantial sensing-modality gap between radar and lidar, which observe and represent scene structure in fundamentally different ways. Dr-LiSA bridges this gap using a learned forward model that predicts radar measurements from a lidar submap at a candidate pose, enabling direct photometric alignment of predicted and observed radar scans in $SE(3)$. Dr-LiSA outperforms prior radar-lidar approaches in $SE(2)$ while achieving planar accuracy competitive with state-of-the-art radar-radar localization across more than 90 km of on-road data.