Radar-to-Lidar: Heterogeneous Place Recognition via Joint Learning
This addresses the challenge of long-term place recognition for robotics and autonomous systems in environments where single-sensor methods fail, though it is incremental as it builds on existing heterogeneous sensor fusion approaches.
The paper tackles the problem of place recognition in adverse conditions by proposing a framework that retrieves query radar scans from existing lidar maps using a deep neural network with joint training, achieving multiple recognition types (lidar-to-lidar, radar-to-radar, and radar-to-lidar) with a single trained model.
Place recognition is critical for both offline mapping and online localization. However, current single-sensor based place recognition still remains challenging in adverse conditions. In this paper, a heterogeneous measurements based framework is proposed for long-term place recognition, which retrieves the query radar scans from the existing lidar maps. To achieve this, a deep neural network is built with joint training in the learning stage, and then in the testing stage, shared embeddings of radar and lidar are extracted for heterogeneous place recognition. To validate the effectiveness of the proposed method, we conduct tests and generalization experiments on the multi-session public datasets compared to other competitive methods. The experimental results indicate that our model is able to perform multiple place recognitions: lidar-to-lidar, radar-to-radar and radar-to-lidar, while the learned model is trained only once. We also release the source code publicly: https://github.com/ZJUYH/radar-to-lidar-place-recognition.