DiSCo-SLAM: Distributed Multi-Robot LiDAR SLAM With Two-Stage Global-Local Graph Optimization

A novel framework is proposed for distributed,multi-robot SLAM intended for use with 3D LiDAR observations. The framework, DiSCo-SLAM, is the first to use the lightweight Scan Context descriptor for multi-robot SLAM, permitting a data-efficient exchange of LiDAR observations among robots. Additionally, this framework includes a two-stage global and local optimization framework for distributed multi-robot SLAM which provides stable localization results that are resilient to the unknown initial conditions that typify the search for inter-robot loop closures. This proposed framework is compared with the widely used distributed Gauss-Seidel (DGS) approach, over a variety of multi-robot datasets, quantitatively demonstrating its accuracy, stability, and data-efficiency.

[yuqi-dp]

What GPU are you using, when I run this I can't get a clear display of the point cloud in rviz, it will appear like a transient, is it a lack of performance in the GPU, I'm using a R7 7700 and it's core graphics, if it's a GPU problem what GPU should I replace it with

[BobBob-sb]

Super cool dude! What software did you use to perform this in? I've been working with Gazebo but really like this visualization.