An Ultra-Wideband-based Multi-UAV Localization System in GPS-denied environments

This paper presents a technique for multiUAV localization using ranging measurements from Two-Way Time-Of-Flight Ultra-Wideband (UWB) transceivers. In continuation with our previous work, the use of the Extended Kalman Filter (EKF) estimate is extended by fusing with other sensors to achieve a usable altitude estimate. Besides EKF, another method based on Non-linear Regression (NLR) is also developed to serve as an auxiliary localization to supplement the EKF. Experiments of autonomous flights are carried out to study the performance of these localization methods. Our success in supporting 4 UAVs demonstrates the capability of UWB for multi-UAV localization as a substitute for GPS in GPS-denied conditions. It can also serve as a cost-competitive alternative for very accurate yet expensive, highly centralized motion capture systems for indoor localization.