Reproduce “GOFusion: Real-Time Gaussian-Occupancy Submap Fusion for Fisher-Informed Autonomous Robot Navigation” 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
Boxin Bu, Simon Boche, Haoming Zhang, Jaehyung Jung, Jiaxin Wei, Hannah Schieber, Daniel Roth, Angela P. Schoellig and 1 more
GOFusion combines occupancy submaps with Gaussian primitives to support both safe navigation and appearance mapping. It builds on visual-inertial SLAM, refining camera poses photometrically while submap pose corrections move both layers together. Its active-submap optimization limits trainable memory. On uHumans2’s 49.4 m apartment sequence, it reports PSNR 33.09 and LPIPS 0.062 across 10 runs; the real flight used offboard mapping.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.