Main Interest Topics
(Deep) Reinforcement Learning ((D)RL)
Multi-agent systems and related learning/execution paradigms:
fully centralized with a base station acting as a main supervisor/controller
fully decentralized
centralized training and decentralized execution mainly based on an actor-critic neural network structure
shared parameter training
Delayed Markovian Decision Processes with Mixed Observability (D-MOMDP) solved through (D)RL:
observation delays
action delays
Path Planning (mainly combined with (D)RL)
Main Application Topics
(Deep) Reinforcement Learning
Multiple Unmanned Aerial Vehicle (Multi-UAV) systems:
homogeneous systems (e.g., only quadcopters UAVs)
cooperation
area coverage
communication maximization through smart services provision
desired target spotting
desired target tracking
Optimization algorithms (e.g., Symplex), Supervised Learning, Deep Learning
Multiple Unmanned Aerial Systems (Multi-UASs) in the U-Space:
heterogeneous systems (e.g., fixed wings and quadcopters UAVs)
trajectory conflicts analysis
trajectory conflicts estimation
deployment optimization based on specific separation minima requirements
flight level assignment
Work
Universities
Projects