DeepPointMap: Advancing LiDAR SLAM with Unified Neural Descriptors
This work addresses a key bottleneck in LiDAR SLAM for robotics and autonomous systems, offering a novel solution to improve efficiency and accuracy, though it appears incremental in advancing existing neural descriptor methods.
The paper tackled the conflict between localization accuracy and map size in LiDAR SLAM by proposing DeepPointMap, a unified architecture that uses neural descriptors to achieve memory-efficient maps and accurate multi-scale localization, with results showing promising performance in multi-agent collaborative SLAM.
Point clouds have shown significant potential in various domains, including Simultaneous Localization and Mapping (SLAM). However, existing approaches either rely on dense point clouds to achieve high localization accuracy or use generalized descriptors to reduce map size. Unfortunately, these two aspects seem to conflict with each other. To address this limitation, we propose a unified architecture, DeepPointMap, achieving excellent preference on both aspects. We utilize neural network to extract highly representative and sparse neural descriptors from point clouds, enabling memory-efficient map representation and accurate multi-scale localization tasks (e.g., odometry and loop-closure). Moreover, we showcase the versatility of our framework by extending it to more challenging multi-agent collaborative SLAM. The promising results obtained in these scenarios further emphasize the effectiveness and potential of our approach.