In this paper, we consider the unmanned aerial vehicles (UAVs) as mobile sensors with flying ability to provide a timely and accurate vision over a large supervised area. The interconnection of these UAVs, in order to route the collected data to the control station on the ground, constitutes the flying ad hoc network (FANET). However, the UAVs features impose challenging issues to data collection and routing. Thus, the communications are prone to high-frequency link disconnections caused by the high mobility of UAVs. In this paper, we propose a reliable architecture to collect and route data to a ground station (GS). The proposed simple area division cluster-based algorithm (SAD-CA) provides a stable network architecture through the clustering technique in order to deal with the high mobility of the UAVs while preserving their energy consumption. The performance evaluation of the proposed algorithm, carried out using the network simulator 3 (NS3), has shown convincing results, in terms of energy consumption and network lifetime, outperforming those exhibited by the referred protocol.