This paper presents a set-membership approach for the coordinated control of a fleet of UAVs aiming to search and track an a priori unknown number of targets spread over some area of interest. The originality of the approach lies in the description of the perturbations and measurement uncertainties via bounded sets. Sets guaranteed to contain the actual state of already detected targets are provided using the set-membership approach. A specific criterion evaluating the uncertainty of the detected and still to be detected target has been defined, which is used to determine the trajectories of the UAVs so as to decrease the global estimation uncertainty. The criterion accounts for potential occlusions in the field of view of UAVs due to obstacles located on the zone. Simulations show that the proposed control input optimization is able to provide good localization and tracking performance for multiple targets.