A novel cooperative searching architecture for multi-unmanned aerial vehicles under restricted communication

Searching for a distinguishable lost target in a bounded area automatically with unmanned aerial vehicle (UAV) is a fundamental problem in the theory of physical search. This paper studies the problem in detail, presents a new and more realistic Bayesian formula representing the problem by taking communication capability of UAVs into consideration, and focuses on the case of applying a team of multiple cooperative UAVs in the field for the search, where each UAV can autonomously search the area for the target using Bayesian filtering algorithm. The perturbation of the searching performance by the different prior belief probability distribution functions is investigated. Empirical evidence findings illustrate that our approach yields improved accuracy.