Context: Unmanned Aerial Vehicles (UAVs) equipped with remote sensing platforms have become increasingly popular due to their applications in aerial surveillance, environmental control, and disaster response. However, the limited flight range and on-board energy resources of UAVs pose significant challenges to their practical deployment and operating efficiency, which has led to the exploration of energy-efficient platforms for remote sensing. Method: This paper proposes a preliminary approach for UAV multi-sensor reconnaissance and surveillance platforms (MRSS) that target low energy consumption. The approach implemented four sensor modules controlled by one multi-functional integrated edge computer for control and data collection, which can be interchanged according to battery lifetime requirements. Results: The main contribution of this work was an analysis of the energy consumption behavior of sensor modules managed by an embedded system with edge computing capabilities as the central control unit. Conclusions: The high energy consumption associated with modules such as GEOINT leads to deep discharge in excess of 20 % DOD, resulting in a maximum battery degradation of 2,4 years.
A current challenge for computer users is to fully exploit performance of new Multicore systems. We propose a methodology for students in computational science to analyze the effect of memory hierarchy on application performance. The analysis is proposed in a experimental environment consisting of different systems with different configurations of memory hierarchy. New Multicore systems put tremendous pressure on memory hierarchy systems. The pressure is because, unfortunately, the effectiveness of the computing power offered by Multicore is affected by the data communications inter-chip and off-chip to the memory hierarchy, leading to significant problems in performance for many parallel applications. In the scope of computer science, it is important that students understand these problems. This methodology was successfully applied to students, where they acquired a significant improvement in their parallel application metrics assessment as was demonstrated in our evaluation.