MNE-SLAM: Multi-Agent Neural SLAM for Mobile Robots

Figure 1 . We present MNE-SLAM, the first distributed multi-agent collaborative SLAM system with distributed mapping and camera tracking, joint implicit neural scene representation, intra-to-inter loop closure, and multiple submap fusion. Depicted at the bottom, we demonstrate the real-world, large-scale long-corridor scenes (≈1000 m 2 ) with high-accuracy 3D groundtruth mesh. This scene is collected through various industrial laser scans. We display the rendered depth and color image of different type of agents around the corridor. The trajectory of each agent is marked in a unique color for clarity.