[Back]

S. Goel, J. Gabela, G. Retscher, C. Toth, A. Masiero, A. Kealy:
"UWB Cooperative Localization of Pedestrians along a Constrained Building Hallway";
Talk: International Global Navigation Satellite Systems (IGNSS) 2020, Sydney, Australia (invited); 2020-02-05 - 2020-02-07; in: "International Global Navigation Satellite Systems (IGNSS) 2020", (2020), Paper ID 7, 13 pages.

In contrast to outdoor localization that relies primarily on GNSS (Global Navigation Satellite Systems) signals, localization in indoor environments is quite challenging due to the absence of GNSS signals and presence of various objects that reflect and disperse the signals such as Wi-Fi (Wireless Fidelity), Ultra-Wide Band (UWB) or other SOP (Signals of Opportunity) used for localization. Cooperative Localization (CL) has proven to be one of the practical approaches for localization in GNSS denied and challenging environments. This paper analyses the experimental performance of an indoor cooperative pedestrian localization approach that relies on UWB based relative range measurements among the pedestrians, as well as relative measurements between the pedestrians and static infrastructure nodes. The experimental setup uses a network containing about 30 static infrastructure nodes and four pedestrians moving in a hallway. Each infrastructure node and pedestrian are equipped with a UWB sensor for relative range measurements. Additionally, some of the pedestrians also carry a low-cost inertial sensor, but inertial observations are not used in this paper. A centralized extended Kalman filter (EKF) is used to localize all the pedestrians using the relative range information and knowledge about the infrastructure nodes. It is observed that the network geometry perceived by a pedestrian has a significant impact on its localization accuracy. The results demonstrate that the proposed approach can achieve decimetre level localization accuracies, provided a good network geometry, and enough range observations are available. Towards either end of the hallway, significant degradation in the localization accuracy is observed due to the combined effect of the decrease in the number of available range observations and poor network geometry. The accuracy achieved in the proposed setup may be further improved by the inclusion of inertial sensor observations and precise time synchronization among the nodes.

Cooperative Localization, Extended Kalman Filter, Indoor Positioning, Ultra-Wide Band