Tightly Coupled SLAM with Imprecise Architectural Plans
🤖 AI Summary
To address geometric misalignments (e.g., translation and rotation) between architectural floor plans and real indoor environments—causing degradation in robot localization and mapping accuracy—this paper proposes a tightly coupled LiDAR SLAM framework for mapping and localization using imprecise floor plans. Our method jointly optimizes robot poses, global floor-plan alignment parameters, and structure-level geometric deviations (translation and rotation) within a unified nonlinear optimization framework, enabling real-time modeling of plan-to-reality discrepancies. We further introduce a multi-layer semantic map representation that integrates geometric and semantic constraints to enhance structural consistency. Experimental results demonstrate a 43% reduction in simulated localization error and a 7% decrease in average 3D map alignment error on real-world data, while maintaining robustness against floor-plan inaccuracies of up to 35 cm translation and 15° rotation.
Application Category
📝 Abstract
Robots navigating indoor environments often have access to architectural plans, which can serve as prior knowledge to enhance their localization and mapping capabilities. While some SLAM algorithms leverage these plans for global localization in real-world environments, they typically overlook a critical challenge: the"as-planned"architectural designs frequently deviate from the"as-built"real-world environments. To address this gap, we present a novel algorithm that tightly couples LIDAR-based simultaneous localization and mapping with architectural plans under the presence of deviations. Our method utilizes a multi-layered semantic representation to not only localize the robot, but also to estimate global alignment and structural deviations between"as-planned"and as-built environments in real-time. To validate our approach, we performed experiments in simulated and real datasets demonstrating robustness to structural deviations up to 35 cm and 15 degrees. On average, our method achieves 43% less localization error than baselines in simulated environments, while in real environments, the as-built 3D maps show 7% lower average alignment error
Problem
Research questions and friction points this paper is trying to address.
Robotics
Indoor Localization
Map Building
Innovation
Methods, ideas, or system contributions that make the work stand out.
LIDAR-based SLAM
Multi-layered Semantic Representation
As-planned vs As-built Alignment
🔎 Similar Papers
No similar papers found.
