Real time multi robot 3D localization system using trilateration

Abstract

This paper presents a three dimensional localization of an unmanned aerial vehicle implemented in real time. The localization is carried on by means of trilateration using a moving formation of omnidirectional robots as stations, and an aerial vehicle as the receiver. The mathematical tools used are explained. The propagation of errors is calculated from measurements in the omnidirectional robots and cameras to the aerial vehicle. Real time experiments are carried on using a motion camera system as a ground truth reference. The algorithm is tested with two estimation techniques: an extended Kalman filter and a particle filter, their differences are discussed.