DeepRING: Learning Roto-translation Invariant Representation for LiDAR based Place Recognition
This addresses a key challenge in robotics for loop closure detection and re-localization, though it appears incremental as it builds on existing deep learning approaches.
The paper tackles the problem of LiDAR-based place recognition degrading under large perspective differences by proposing DeepRING, which learns roto-translation invariant representations, and shows it outperforms comparative methods with improved dataset-level generalization.
LiDAR based place recognition is popular for loop closure detection and re-localization. In recent years, deep learning brings improvements to place recognition by learnable feature extraction. However, these methods degenerate when the robot re-visits previous places with large perspective difference. To address the challenge, we propose DeepRING to learn the roto-translation invariant representation from LiDAR scan, so that robot visits the same place with different perspective can have similar representations. There are two keys in DeepRING: the feature is extracted from sinogram, and the feature is aggregated by magnitude spectrum. The two steps keeps the final representation with both discrimination and roto-translation invariance. Moreover, we state the place recognition as a one-shot learning problem with each place being a class, leveraging relation learning to build representation similarity. Substantial experiments are carried out on public datasets, validating the effectiveness of each proposed component, and showing that DeepRING outperforms the comparative methods, especially in dataset level generalization.