Cooperative positioning algorithm of swarm UAVs based on posterior linearization belief propagation

Aiming at the problem of cooperative localization of swarm UAVs in complex flight environment, a cooperative positioning algorithm of swarm UAVs based on posterior linearization belief propagation is proposed. According to the belief propagation algorithm, the cooperative localization of drones using belief propagation is proposed. To deal with the contradiction between positioning accuracy and computational complexity, statistical linear regression is used to design the posterior linearization method of measurement model, and the linearized measurement model is used on the algorithm framework of cooperative positioning which is based on belief propagation. Finally, simulation comparison analysis is carried out for different algorithms. The simulation results show that, compared with the non-cooperative case, the proposed algorithm improves the positioning accuracy of the follower UAVs in X, Y and Z directions (earth-centered earth-fixed coordinate system, ECEF) by 8.9, 13.9, 17.3 times respectively.

I. Introduction

At present, the multi-UAVs formation technology has developed rapidly, and it has great potential application value in the fields of battlefield reconnaissance, communication relay, and 3D reconstruction. Due to the small distance between the swarm UAVs and the high dynamics during flight, the high-precision and high-reliability positioning and navigation information is an important basis for the swarm UAVs to perform tasks efficiently [1].