Volume 17, Number 6

RLSUAV: Relative Localization in a Swarm of UAVS

  Authors

Sara Benkouider, Nasreddine Lagraa and Mohamed Bachir Yagoubi, Université Amar Telidji de Laghouat, Algeria

  Abstract

A swarm or fleet of Unmanned Aerial Vehicles (UAVs) can be used to accomplish several missions such as security, search and rescue or surveillance in unknown and dangerous environments. In order to deploy a fleet of drones for such applications, drones must be able to perform certain tasks, such as collision prevention, and formation flight with a leader node. These tasks are accomplished by knowing the location of neighboring drones in the group. The conventional method of determining the position relies mainly on the GPS system. Therefore, drone swarms relying on classical positioning methods (GPS) cannot operate in dense urban or in indoor environments, due to the difficulties encountered in receiving the GPS signal from the satellites. In these cases, relative localization can be used to help nodes without GPS to determine their positions. Relative localization uses cooperative communication and information sharing between nodes in the network to help them estimate their positions. In this paper, we propose a novel technique providing relative localization in a swarm of UAVs (RLSUAV) that does not require any GPS information and, therefore, can be used in indoor environments. It consists of estimating the positions of the drone nodes as a function of the distances measured between them, combined with the multilateration technique, where the distances between the drones are calculated using the power of the received signal (RSSI). Simulation results showed the effectiveness of RLSUAV in different environments (with and without multipath), achieving an estimation error inferior to 5 meters in most cases.

  Keywords

UAV; Swarm; Relative localization; GPS; Indoor environment; Multilateration; RSSI; Multipath