Simultaneous Localization and Mapping Using the Phase of Passive UHF-RFID Signals

Robot Localisation Using UHF-RFID Tags: A Kalman Smoother Approach †

“…It is worth mentioning the work presented in [ 18 ], where an indoor simultaneous localisation and mapping (SLAM) problem is considered. A set of passive tags was deployed at the ceiling in unknown locations.…”

“…The second one, on the contrary, is based on having the vehicle equipped with a reader connected to one or more antennas, while a grid of RFID tags is deployed in the surrounding scenario at known locations [ 17 ]. In case of unknown reference tag positions, simultaneous localisation and mapping (SLAM) systems has been also designed [ 18 ]. For sure, the second class of solutions is arguably more appealing to a possible implementer since the passive UHF-RFID tags do not require direct power supply (contrary to the UWB nodes).…”

“…The data association can be easily performed thanks to the tag unique identifier (Electronic Product Code, EPC). Moreover, UHF-RFID tags can be easily installed at the floor side [ 19 ], at the ceiling [ 18 ] or in almost any other place in the considered environment [ 20 ]. In harsh industrial scenarios with lots of metallic structures, Commercial-Off-The-Shelf (COTS) on metal tags can be employed.…”

A Revisiting Method Using a Covariance Traveling Salesman Problem Algorithm for Landmark-Based Simultaneous Localization and Mapping

“…Rao–Blackwellized particle filters (RBPFs) are appropriate for landmark-based maps [8], and also efficiently optimize gridmaps and estimate robot poses [9,10]. Radio frequency identification (RFID) phase measurements were employed to estimate the robot pose and passive tag coordinates by using the landmark-based RBPF SLAM approach [11]. Whereas most previous gridmap-based RBPF SLAM algorithms have modeled each cell as one Bernoulli distribution [12,13], the occupancy probability of each cell is modeled by using a beta-distribution in β-SLAM to distinguish between different causes of uncertainty [14].…”

“…Most loop-closing strategies have focused on detection of loop-closing landmarks and loop optimization [21,26,27]. Several approaches simplify loop-detection issues using special landmarks [11,28]. Appropriate loop-closing positions and planning of loop-closing paths have received less attention.…”