Research Activity
My long-term research goal is to develop efficient approaches to enable multi-robot systems to autonomously operate in real-world environments.
This is a very ambitious goal, that requires significant advances both in Artificial Intelligence and in Robotics. My contribution to reach this goal consists of researching new algorithmic solutions which allow artificial agents to develop skills, learn solutions of complex tasks, and coordinate their activities.
Up to now, my main achievements have been in the following research areas:
Reinforcement Learning
LEAP: an approach for learning in large state applications
SMILe: a framework that allows an agent to autonomously develop general-purpose skills driven by self-motivation
Multi-agent systems
SCARE: an architecture for heterogeneous agent coordination
CoLF: an algorithm for improving learning in multi-agent scenarios
Robotics
MRT: a team of soccer robots, participating at the RoboCup competition
MUREA: a localization algorithm based on evidence accumulation