Evolutionary Cloud for Cooperative UAV Coordination

摘要

This report is dedicated to considerations and perspectives for the strategy of using unmanned aerial vehicles and Cloud Computing in emergency situa-tions in a Smart City environment. The aim is to provide inspiring insights in this highly complex problem from a technological and a tactical point of view. This specific case can be seen as a member of a family of similar problems, and hence the solutions proposed in this chapter can be used to solve a large set of related cases. The efficient use of the presented set-up would require to solve a time-consuming multi-objective optimization problem, which one could- at least partially- solve using some form of Evolutionary Computing. Emergency situations, by default, are unpredictable and often involves in-complete information which might be supplemented over time. The changing situation creates an unpredictable complexity for the computational problem and therefore we will need to cope with bursts in the need for computational power. To overcome this problem we propose to perform the computation us-ing Cloud Computing. This solution also improves the robustness of the overall system because it provides fail-over and content replication. In order to inte-grate and analyze the various measurements performed by the robots we sug-gest the use of Semantic Web technologies. However, the size of the sensing data can be grow to enormous sizes and we note that this problem can be con-sidered as a Big Semantic Data problem. Based on our earlier work, we propose to tackle this problem using an Evolving Knowledge Ecosystem. 1