Chan, C. L., Li, J., Chan, J. L., Li, Z., & Wan, K. W. (2021). Partial-Map-Based Monte Carlo Localization in Architectural Floor Plans. Lecture Notes in Computer Science, 541–552. doi:10.1007/978-3-030-90525-5_47
Abstract:
Mobile robots, in modern technology, demand a more ro-bust localization in a complex environment. Currently, the most com-monly used 2D LiDAR localization system for mobile robots requiresmaps that are constructed by 2D SLAM.Such systems do not cope wellwith dynamic environments and also have high deployment costs whenmoving robots to a new environment setting as they require the recon-struction of a map for each new place. In modern days, a floor plan isindispensable for an indoor environment. It typically represents essentialstructures such as walls, corners, pillars, etc. for humans to navigate inthe environment. This information turns out to be crucial for robot local-ization. In this paper, we propose an approach for 2D LiDAR localizationin an architectural floor plan. We use partial simultaneous localizationand mapping (PSLAM) algorithm to generate a map while we concur-rently aligned it to the floor plan using Monte Carlo Localization (MCL)method. Real-world experiments have been conducted with our proposedmethod which results in robust robot localization, the algorithm is evenevaluated on a large discrepancies floor plan (discrepancies between thefloor plan and real-world). Our algorithm demonstrates that its capabil-ities of localizing in real-time applications
License type:
Publisher Copyright
Funding Info:
This research / project is supported by the National Robotics Programme (NRP) - SERC Grant
Grant Reference no. : 192 25 00049