CFEAR-Teach-And-Repeat: Fast and Accurate Radar-Only Localization
Maximilian Hilger, Daniel Adolfsson, Ralf Becker, Henrik Andreasson, Achim J. Lilienthal
AI summary
Problem
Existing radar-only localization methods lag behind lidar in accuracy and heading estimation, while radar-inertial systems require complex calibration and synchronization. This limits simple, robust deployment in GPS-denied or adverse weather conditions.
Approach
The system uses a teach-and-repeat framework that jointly aligns live radar scans to both a prior map and a sliding window of recent live keyframes via sparse oriented surface points, eliminating the need for IMU fusion or multi-sensor calibration.
Key results
- 0.117 m translation and 0.096° heading error on held-out Boreas test sequences
- Up to 63% improvement in heading estimation over prior radar-only state of the art
- Efficient real-time operation at 29 Hz without additional sensors
- State-of-the-art radar-only odometry performance with 0.42% translation drift
Why it matters
Enables robust, low-cost autonomous navigation in GPS-denied or adverse weather environments where optical sensors fail.
Abstract
Reliable localization in prior maps is essential for autonomous navigation, particularly under adverse weather, where optical sensors may fail. We present CFEAR-TR, a teach-and-repeat localization pipeline using a single spinning radar, which is designed for easily deployable, lightweight, and robust navigation in adverse conditions. Our method localizes by jointly aligning live scans to both stored scans from the teach mapping pass, and to a sliding window of recent live keyframes. This ensures accurate and robust pose estimation across different seasons and weather phenomena. Radar scans are represented using a sparse set of oriented surface points, computed from Doppler-compensated measurements. The map is stored in a pose graph that is traversed during localization. Experiments on the held-out test sequences from the Boreas dataset show that CFEAR-TR can localize with an accuracy as low as 0.117 m and 0.096°, corresponding to improvements of up to 63% over the previous state of the art, while running efficiently at 29 Hz. These results substantially narrow the gap to lidar-level localization, particularly in heading estimation. We make the C++ implementation of our work available to the community.