[Back]

S. Goel, A. Kealy, G. Retscher, B. Lohani:
"Cooperative P2I Localization Using UWB and Wi-Fi";
Talk: IGNSS 2016 Conference, Sydney, Australien (invited); 2016-12-06 - 2016-12-08; in: "IGNSS 2016", (2016), Paper ID 16, 16 pages.

A Peer to Infrastructure (P2I) cooperative localization system makes use of existing infrastructure network for localization of mobile platforms. This interconnected network can be utilized for several applications besides communication such as disaster management, situational awareness, search and rescue, guided navigation, Cooperative Intelligent Transport System etc. The focus of this paper is development and analysis of a localization system using a well-established infrastructure network. This paper develops methods of localization of dynamic platforms such as Unmanned Aerial Vehicles (UAV) equipped with GNSS, inertial sensors (INS), ultra-wide band (UWB) and wireless local area network (Wi-Fi) receivers. Traditional platforms (cars, UAVs etc.) rely on GNSS for accurate localization and thus cannot be used in indoor environments or other GNSS challenging regions. This paper leverages continuously deployed networks of urban infrastructure that provide signals of opportunity such as Wi-Fi, as well as UWB to develop a robust localization system which is capable of providing a continuous and seamless position solution in both indoor and outdoor environments. This paper also provides an analysis of this localization system using numerical simulations and discusses experimental results of a developed prototype. The contributions of this paper include development and analysis of a prototype cooperative localization system and quantification of its performance. Initial simulation results show that in GNSS denied regions, a platform may be able to achieve positioning accuracy comparable to the accuracy provided by a GNSS under certain conditions. In GNSS available regions, other signals further improve the solution obtained using only GNSS. The simulation results are validated using experiments and the developed system achieves an accuracy of better than 10m more than 95% of the time using a maximum of 4 UWB measurements when GNSS is not present at all and more than 3 UWB measurements are available only 2.5% of time. Higher localization accuracy can be achieved using dense infrastructure network and better quality of UWB measurements.

Cooperative Localization, Extended Kalman Filter, Ultra-Wide Band (UWB), Multi sensor fusion