Autonomous Exploration and Semantic Updating of Large-Scale Indoor Environments with Mobile Robots

2024-09-23

arXiv.org

AI Summary

To address the challenges of semantic mapping and dynamic map updating for mobile robots in unknown indoor environments, this paper proposes a hybrid representation integrating geometric occupancy grids with topological semantic maps. A 2D occupancy grid, constructed from LiDAR data, provides metric grounding; RGB-D sensing enables object detection and semantic classification; and a dynamically modifiable topological graph—comprising insertable/deletable nodes—efficiently encodes and updates semantic entities (e.g., movable objects). The framework supports fully autonomous exploration, real-time semantic mapping, and online map correction under change detection. Extensive validation was conducted on a Fetch robot platform across two distinct environments: a large open space (93 m × 90 m) and a compact scene (9 m × 13 m). Results demonstrate significantly improved semantic map update accuracy and, for the first time, achieve lightweight, topology-driven dynamic semantic updates in large-scale indoor settings.

Technology Category

• Intelligent Robots: Localization, Mapping, and Navigation
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Planning, Routing, and Scheduling: Mixed Discrete/Continuous Planning

Application Category

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

Abstract

We introduce a new robotic system that enables a mobile robot to autonomously explore an unknown environment, build a semantic map of the environment, and subsequently update the semantic map to reflect environment changes, such as location changes of objects. Our system leverages a LiDAR scanner for 2D occupancy grid mapping and an RGB-D camera for object perception. We introduce a semantic map representation that combines a 2D occupancy grid map for geometry with a topological map for object semantics. This map representation enables us to effectively update the semantics by deleting or adding nodes to the topological map. Our system has been tested on a Fetch robot, semantically mapping a 93m x 90m and a 9m x 13m indoor environment and updating their semantic maps once objects are moved in the environments

Problem

• Autonomous exploration and mapping of unknown indoor environments.
• Semantic map updating to reflect environmental changes like object movements.
• Integration of LiDAR and RGB-D cameras for geometry and object perception.

Innovation

• LiDAR and RGB-D for mapping and perception
• Combined 2D grid and topological semantic map
• Dynamic semantic map updating via node manipulation