Active Neural Mapping at Scale

TL;DR

This paper presents a scalable active mapping system using NeRF and GVG for efficient exploration and reconstruction of large indoor environments, integrating geometry, appearance, and uncertainty for adaptive exploration.

Contribution

It introduces a novel NeRF-based active mapping approach that extracts a GVG for adaptive exploration, enabling efficient large-scale indoor environment mapping.

Findings

Achieves high reconstruction accuracy and coverage in large environments

Demonstrates efficient exploration with adaptive path planning

Validates effectiveness across various scales

Abstract

We introduce a NeRF-based active mapping system that enables efficient and robust exploration of large-scale indoor environments. The key to our approach is the extraction of a generalized Voronoi graph (GVG) from the continually updated neural map, leading to the synergistic integration of scene geometry, appearance, topology, and uncertainty. Anchoring uncertain areas induced by the neural map to the vertices of GVG allows the exploration to undergo adaptive granularity along a safe path that traverses unknown areas efficiently. Harnessing a modern hybrid NeRF representation, the proposed system achieves competitive results in terms of reconstruction accuracy, coverage completeness, and exploration efficiency even when scaling up to large indoor environments. Extensive results at different scales validate the efficacy of the proposed system.