LITE: A Learning-Integrated Topological Explorer for Multi-Floor Indoor Environments

2025-07-29

AI Summary

Autonomous exploration in multi-floor indoor environments remains an open challenge, with existing learning-based approaches largely confined to 2D settings and ill-suited for 3D cross-floor navigation. Method: We propose LITE—a hybrid framework integrating learning-based exploration with topological modeling. It first constructs a floor-staircase topological graph, embedding 2D exploration policies into topological nodes; employs attention mechanisms to capture inter-region spatial dependencies; and deploys a finite-state machine to orchestrate cross-floor transitions. Real-time semantic segmentation via YOLOv11 and incremental topological updates enable dynamic environment modeling and adaptive decision-making. Contribution/Results: LITE achieves significant performance gains over state-of-the-art baselines on HM3D and MP3D benchmarks. It is further validated on a quadrupedal robot platform, demonstrating strong generalization across unseen buildings and plug-and-play compatibility with diverse robotic systems.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Search and Optimization: Learning to Search
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

This work focuses on multi-floor indoor exploration, which remains an open area of research. Compared to traditional methods, recent learning-based explorers have demonstrated significant potential due to their robust environmental learning and modeling capabilities, but most are restricted to 2D environments. In this paper, we proposed a learning-integrated topological explorer, LITE, for multi-floor indoor environments. LITE decomposes the environment into a floor-stair topology, enabling seamless integration of learning or non-learning-based 2D exploration methods for 3D exploration. As we incrementally build floor-stair topology in exploration using YOLO11-based instance segmentation model, the agent can transition between floors through a finite state machine. Additionally, we implement an attention-based 2D exploration policy that utilizes an attention mechanism to capture spatial dependencies between different regions, thereby determining the next global goal for more efficient exploration. Extensive comparison and ablation studies conducted on the HM3D and MP3D datasets demonstrate that our proposed 2D exploration policy significantly outperforms all baseline explorers in terms of exploration efficiency. Furthermore, experiments in several 3D multi-floor environments indicate that our framework is compatible with various 2D exploration methods, facilitating effective multi-floor indoor exploration. Finally, we validate our method in the real world with a quadruped robot, highlighting its strong generalization capabilities.

Problem

• Develops a 3D explorer for multi-floor indoor environments
• Integrates learning-based 2D methods into floor-stair topology
• Enhances exploration efficiency using attention-based spatial dependencies

Innovation

• Decomposes environment into floor-stair topology
• Uses YOLO11-based instance segmentation model
• Implements attention-based 2D exploration policy