Enhanced indoor positioning through human-robot collaboration

Abstract

AbstractIndoor positioning is a critical component for numerous applications and services. However, GNSS systems face challenges in delivering accurate positioning information in indoor environments. Current indoor positioning research primarily concentrates on enhancing the positioning performance of individual terminals through various techniques. As we transition into the Internet of Things (IoT) era, former indoor positioning methods need refinement. In this paper, we propose a novel indoor positioning method that leverages robots as mobile base stations to mitigate the problem of inadequate fixed base stations and aims to enhance positioning accuracy by incorporating pedestrian inertial navigation data. The process involves several steps. First, the mobile robots accurately determine their positions and performing coordinate transformations to ensure consistency with pedestrian coordinate systems. Then, pedestrians use the ranging information from these robots along with their smartphones’ sensors for multi-source fusion positioning. Finally, an Extended Kalman Filter (EKF) is applied to fuse the multiple sources of data, considering various sources of errors, to provide enhanced positioning performance. Experimental results demonstrate the effectiveness of this approach in addressing indoor positioning challenges. This method could benefit numerous scenarios involving robots, enhancing pedestrian positioning accuracy and overall system robustness. The paper provides a comprehensive exploration of this proposed method, its implications, and potential directions for future advancements.