GOFusion enables real-time robot mapping and navigation with Gaussian maps
GOFusion: Real-Time Gaussian-Occupancy Submap Fusion for Fisher-Informed Autonomous Robot Navigation
Robotics
Summary
Robots need to understand their surroundings to move safely and efficiently. The authors developed GOFusion, a system that uses special 3D shapes called Gaussian primitives to create detailed maps from a robot’s cameras and sensors in real time. This method helps the robot plan paths and correct its position while accounting for uncertainty. They also introduced a way to prioritize important map details to guide exploration better. Their system works well on public datasets and real drone flights, enabling autonomous navigation.
What this means in practice
- •For robotics engineers: Enable drones and robots to build detailed, uncertainty-aware 3D maps in real time for better navigation and autonomous exploration.
- •For drone operators: Support autonomous drone flights with improved map quality and navigation that adapts to sensor uncertainty and environment complexity.
Authors
Boxin Bu, Simon Boche, Haoming Zhang, Jaehyung Jung, Jiaxin Wei, Hannah Schieber, Daniel Roth, Angela P. Schoellig, Stefan Leutenegger
Abstract
In this letter, we present GOFusion, a real-time visual mapping system for autonomous robot navigation that anchors Gaussian primitives to local occupancy maps, providing both spatial awareness for path planning and high-fidelity appearance that enables information-driven robot navigation. GOFusion builds 3D Gaussian maps from pose estimates and local occupancy maps obtained via visual-inertial SLAM and enables pose corrections to maintain both geometric and visual consistency. Furthermore, it restricts the 3D Gaussian Splatting optimization to active submaps to bound GPU memory usage. Building on the appearance-rich Gaussian map, we introduce a rasterizer-integrated Fisher approximation that quantifies per-primitive sensitivity online, serving as an information-driven heuristic for Gaussian pruning and next-best-view planning. We evaluate GOFusion on public datasets and in real-world drone flights. It not only outperforms several baselines in mapping quality, but also achieves real-time, high-fidelity and uncertainty-aware mapping at scale. Finally, we demonstrate fully autonomous drone exploration for Gaussian map refinement using the proposed Fisher-informed heuristic. Code is available at https://github.com/Bennnoo/GOFusion. We provide a supplementary video demonstration at https://youtu.be/pLQKbxQvb6I.