Multi-robot autonomous 3D reconstruction using Gaussian splatting with Semantic guidance
This work addresses the problem of slow and inefficient autonomous 3D reconstruction in large-scale scenes for robotics and computer vision applications, representing a novel multi-robot extension with incremental improvements in planning and quality.
The paper tackles the challenge of rapid, large-scale 3D reconstruction by proposing the first centralized multi-robot framework based on 3D Gaussian splatting, integrating semantic guidance to improve efficiency and quality, achieving the highest reconstruction quality among planning methods and superior planning efficiency compared to existing multi-robot approaches.
Implicit neural representations and 3D Gaussian splatting (3DGS) have shown great potential for scene reconstruction. Recent studies have expanded their applications in autonomous reconstruction through task assignment methods. However, these methods are mainly limited to single robot, and rapid reconstruction of large-scale scenes remains challenging. Additionally, task-driven planning based on surface uncertainty is prone to being trapped in local optima. To this end, we propose the first 3DGS-based centralized multi-robot autonomous 3D reconstruction framework. To further reduce time cost of task generation and improve reconstruction quality, we integrate online open-vocabulary semantic segmentation with surface uncertainty of 3DGS, focusing view sampling on regions with high instance uncertainty. Finally, we develop a multi-robot collaboration strategy with mode and task assignments improving reconstruction quality while ensuring planning efficiency. Our method demonstrates the highest reconstruction quality among all planning methods and superior planning efficiency compared to existing multi-robot methods. We deploy our method on multiple robots, and results show that it can effectively plan view paths and reconstruct scenes with high quality.