Agents
MAGICIAN: Efficient Long-Term Planning with Imagined Gaussians for Active Mapping
arXiv:2603.22650v2 Announce Type: replace Abstract: Active mapping aims to determine how an agent should move to efficiently reconstruct unknown environments. Most existing approaches rely on greedy n
arXiv:2603.22650v2 Announce Type: replace Abstract: Active mapping aims to determine how an agent should move to efficiently reconstruct unknown environments. Most existing approaches rely on greedy next-best-view prediction, resulting in inefficient exploration and incomplete reconstruction. To address this, we introduce MAGICIAN, a novel long-term planning framework that maximizes accumulated surface coverage gain through Imagined Gaussians, a scene representation based on 3D Gaussian Splatting, derived from a pre-trained occupancy network with strong structural priors. This representation enables efficient coverage gain computation for any novel viewpoint via fast volumetric rendering, allowing its integration into a tree-search algorithm for long-horizon planning. We update Imagined Gaussians and refine the trajectory in a closed loop. Our method achieves state-of-the-art performance across indoor and outdoor benchmarks with varying action spaces, highlighting the advantage of long-term planning in active mapping.
Related
- BrainMem: Brain-Inspired Evolving Memory for Embodied Agent Task Planning
- AnyImageNav: Any-View Geometry for Precise Last-Meter Image-Goal Navigation
- Dual-Anchoring: Addressing State Drift in Vision-Language Navigation
- Online Navigation Planning for Long-term Autonomous Operation of Underwater Gliders
- D-BDM: A Direct and Efficient Boundary-Based Occupancy Grid Mapping Framework for LiDARs
Source: arXiv cs.CV | 2026-04-22