On November 8th, Yousef Emami successfully defended his PhD thesis, supervised by Kai Li, and co-supervised by Eduardo Tovar and Mario Sousa, at the Faculty of Engineering of the University of Porto, Portugal.
Developed with CISTER support, his thesis entitled "Deep reinforcement learning for joint cruise control and intelligent data acquisition in UAVs-assisted sensor networks" proposes a solution to optimize velocity controls and data collection schedules of multiple UAVs, using a multi-UAV deep reinforcement learning-based scheduling algorithm (MADRL-SA). This minimizes packet loss even with outdated sensor knowledge. In unmanned aerial vehicle (UAV)-assisted sensor networks (UASNets), swift movements extend the age of information (AoI), and slow movements prolong flight time, affecting ground sensor AoI. Coordinating UAV velocities minimizes average AoI. To adress this issue, Yousef's thesis presents a mean-field flight resource allocation optimization, treated as a mean-field game (MFG), and introduces a mean-field hybrid proximal policy optimization (MF-HPPO) scheme to handle the expanded solution space.
The PhD examination committee was composed of Prof. José Nuno Moura Marques Fidalgo, Associate Professor at FEUP, Prof. Xiaoming Fu, Professor at the Institute of Computer Science, Faculty of Mathematics and Computer Science, University of Goettingen, Prof. José Nuno Panelas Nunes Lau, Associate Professor, Department of Electronics, Telecommunications and Informatics, University of Aveiro, Prof. Pedro Miguel Salgueiro Santos, Invited Assistant Professor, Department of Electrical and Computer Engineering, Faculty of Engineering, University of Porto, Prof. Kai Li, Researcher at the Research Centre in Real-Time and Embedded Computing Systems of Porto's Higher Institute of Engineering of the Polytechnic Institute of Porto and Prof. Maria do Rosário Marques Fernandes Teixeira de Pinho, Associate Professor, Department of Electrical and Computer Engineering, Faculty of Engineering, University of Porto.
Mr. Emami is a passionate inventor and researcher with a Ph.D. from the University of Porto, Portugal. His area of expertise lies in the intersection of wireless communications and artificial intelligence, particularly in the automation of data collection tasks in UAVs-assisted sensor networks, which have resulted in significant contributions in this field. During his doctoral studies at CISTER, Yousef focused on cruise control and data collection scheduling to provide reliability and data freshness in UAVs-assisted sensor networks. He was actively involved in the ARNET project, which aimed at developing innovative solutions to enhance network reliability and security in Airborne Relaying NETworks (ARNET).
