RSAR: A New Method for Accurate Localization of Tagged Objects via RFID-Equipped Robots
Lei Wang, Yongtao Ma, Dianfei Su, Wanru Ning, Kaihua Liu
Abstract
Thanks to the vigorous development of smart logistics and warehouse applications, many scholars have made outstanding contributions to indoor object positioning, among which indoor positioning algorithms mediated by RFID(Radio frequency identification) systems are particularly prominent. However, the phase information of RFID has the characteristics of high precision, the state-of-art algorithm generally uses phase information to achieve localization. Inspired by the SAR(synthetic aperture radar) method that utilizes phase information, this paper first proposes an RSAR(rotating synthetic aperture radar) method for harnessing RSSI(received signal strength indicator) information. This method establishes a functional relationship between RSSI information and angle, and virtually synthesizes reader antenna arrays at different angles at the same point by rotating the reader antenna. By combining the RSAR method using RSSI information with the SAR method using phase information, a 3D positioning system with mobile robots as the carrier is proposed. After that, this paper proposes a position estimation algorithm named SCWT(Sort Cluster Algorithm With Threshold) to improve positioning accuracy. Finally, we evaluate the performance of the three aspects of our proposed algorithm through multiple experiments. First, experiments show that the RSAR method can achieve high-precision positioning. Second, the RSAR method and the SAR method are independent of each other when realizing 3D positioning. Third, comparative experiments show that the time complexity of the proposed position estimation algorithm is more dominant.