Robust Simultaneous Localization and Mapping Using Range and Bearing Estimation of Radio Ultra High Frequency Identification Tags

• Published in: IEEE transactions on control systems technology Vol. 31; no. 2; pp. 1 - 14
• Main Authors: Romanelli, Fabrizio, Martinelli, Francesco, Giampaolo, Emidio Di
• Format: Journal Article
• Language: English
• Published: New York IEEE 01-03-2023; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Backscattering; Coders; Extended Kalman filter; Frequency measurement; High frequency; Localization; Location awareness; Phase measurement; Position measurement; Radio frequency identification; radio frequency identification (RFID) localization; Reliability analysis; Resilient sensor fusion; Robots; robust Kalman filtering; Robustness (mathematics); Simultaneous localization and mapping; simultaneous localization and mapping (SLAM); Tags; ultra high frequency -radio frequency identification (UHF-RFID); Ultrahigh frequencies
• ISSN: 1063-6536; 1558-0865
• DOI: 10.1109/TCST.2022.3204386

In this article, we consider an indoor simultaneous localization and mapping (SLAM) problem for a mobile robot measuring the phase of the signal backscattered by a set of passive radio ultra high frequency identification (ID) tags, deployed in unknown position on the ceiling of the environment. The solution approach is based on the introduction, for each radio frequency identification (RFID) tag observed, of a multihypothesis extended Kalman filter (MHEKF) which, based on the measured phases and on the wheel encoder readings, provides an estimate of the range and of the bearing of the observed tag with respect to the robot. This information is then used in an extended Kalman filter (EKF) solving the SLAM problem. Since an effective range and bearing estimate is available only after some steps, a resilient module is added to the algorithm to evaluate the reliability of the position estimate of each observed tag. As shown through numerical and experimental results, this makes the proposed approach robust with respect to several kinds of unmodeled disturbances, like multipath effects or even the unexpected change of the position of a tag.