Reproduce “MR-SLAM: Immersive Spatial Supervision for Multi-Robot Mapping via Mixed Reality” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Prakash Aryan, Cem Erdogdu, Kavinaya Kumarchokkappan, Timo Kehrer, Sebastiano Panichella

The MR-SLAM system introduces an immersive mixed reality interface for supervising multi-robot Simultaneous Localization and Mapping (SLAM) tasks, integrating passthrough visualization with real-time occupancy grid merging. Utilizing a Meta Quest 3 headset, the system demonstrated an average map coverage of 17.9  0.8 m and maintained 94.7  0.5% cross-instance occupancy consistency across independently operating simulated robots.

Key insights

  • Immersive mixed reality can effectively reduce cognitive load and enhance situational awareness for supervising multi-robot SLAM by directly overlaying robot data onto the physical environment.
  • Consumer-grade MR hardware can be robustly integrated with a ROS 2 backend for complex multi-robot tasks through careful handling of communication challenges such as TF partitioning and QoS mismatches.
  • Consistent and coherent merged occupancy grids can be achieved in real-time from multiple independently operating SLAM instances under immersive human supervision.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.