Azim Akhtarshenas

I am a final-year PhD candidate in Electrical Engineering and Telecommunication at the Universitat Politècnica de València (UPV), working at the iTEAM Research Institute under the supervision of Prof. David López-Pérez.

My PhD develops deep reinforcement learning (PPO, TD3) for positioning and trajectory optimization of UAV and HAPS base stations, including a GPS-free, radio-sensing framework that coordinates multiple UAV base stations in disaster-response networks — where GPS and terrestrial infrastructure may be unavailable — using only angle-of-arrival and signal-quality measurements from the ground users it serves. This centers on autonomous decision-making under uncertainty: agents that learn continuous control policies over 3D positioning and trajectories, and that coordinate as a team rather than act independently, so a fleet of UAV-BSs can maintain coverage as conditions change in real time. More broadly, I work across artificial intelligence and machine learning (federated learning, autoencoders, large language models) and wireless communications (MIMO, 5G/6G, SINR estimation), and I'm increasingly drawn to taking reinforcement learning out of simulation and into real autonomous systems.