Skeleton-oriented Autonomous Reference Point Deployment for Wireless Indoor Localization

With the prosperity of location-based services in recent years, fingerprinting techniques for position estimation have been thriving due to its accuracy and cost-effectiveness compared to other methods. However, the time-consuming and labor-intensive process of data collection on specific reference points in fingerprinting schemes is always the barrier to real world implementation. In this paper, we proposed a novel robot automation system to serve as an alternative method for laborious data collection via human being. We designed the skeleton-oriented autonomous reference point deployment (SARD) system for wireless indoor localization, which integrates floor plan generation, reference point deployment, and data acquisition. The autonomous robot is capable of building the fingerprint database of an unknown space in a single spatial time traversal run. According to the result, compared to human efforts, the proposed SARD system could establish the database in a relatively shorter time period of around 40 minutes and achieve a low average error distance of 1.39 meters in position estimation.