Spinning radar technique improves place recognition in autonomous vehicles
Automotive mmWave Spinning Radar Place Recognition with Spatially Gated Feature-Correlation Representation
RoboticsArtificial IntelligenceComputer Vision and Pattern Recognition
Summary
Recognizing places you've been before is important for self-driving cars to know where they are. Using spinning radar sensors that see all around the vehicle helps even in bad weather or darkness. The researchers designed a new method called SGCA-Net that looks closely at radar data to recognize places more accurately, even when the car is facing different directions. Tests show that this method works better than previous ones and can adapt to new environments and radar types without extra training.
What this means in practice
- •For autonomous vehicle engineers: Improve vehicle localization by recognizing previously visited places using robust radar data processing under different headings and environments.
- •For robotics navigation teams: Enhance robot re-localization in outdoor environments by applying radar-based place recognition that works across weather and lighting conditions.
Authors
Saimunur Rahman, Sagun Singh Shrestha, Abdelwahed Khamis, Peyman Moghadam
Abstract
Automotive spinning FMCW radar provides dense, $360^\circ$ sensing and remains reliable under poor illumination and adverse weather, making it well-suited to autonomous navigation. Place recognition uses these observations to identify previously visited locations for re-localization and long-term navigation. However, heading changes appear as circular shifts in the polar radar representation, and conventional global aggregation can lose relationships among radar responses that are important for distinguishing similar places. We propose SGCA-Net, a spinning radar place recognition framework that combines rotation-robust feature extraction with Spatially Gated Correlation Aggregation (SGCA). SGCA learns spatial weights to reduce the influence of unstable and ambiguous radar regions, while aggregating pairwise correlations among local responses to preserve informative feature relationships. Experiments on the MulRan dataset show that SGCA-Net consistently outperforms SOTA methods across urban, campus, and open-road environments, while remaining robust to substantial heading variation. Evaluation on the HeRCULES dataset further demonstrates that SGCA-Net generalizes to unseen environments and radar sensors without fine-tuning.