Stefano Covone

PhD student

Short Bio

Stefano Covone is currently a Ph.D. student in Modeling and Engineering Risk and Complexity at Scuola Superiore Meridionale.

He graduated in Automation and Robotics Engineering in 2024 at the University of Naples Federico II, with a thesis on "Learning-based control of the multi-agent shepherding problem" under the supervision of Prof. Mario di Bernardo.

His current research focuses on developing learning-based control strategies for search and rescue operations in multi-agent systems. His main research interests lie in the study of emergent behaviors in multi-agent systems through learning-based control approaches. His Ph.D. project, titled "Harnessing complex systems for Control: a learning-based control approach to multi-robot shepherding problems in search & rescue operations," is supervised by Prof. Mario di Bernardo (University of Naples Federico II), Dr. Francesco De Lellis (University of Naples Federico II) and Prof. Mirco Musolesi (University College London).

Publications

Preprints:

• S. Covone, I. Napolitano, F. De Lellis, M. di Bernardo, “Hierarchical Policy-Gradient Reinforcement Learning for Multi-Agent Shepherding Control of Non-Cohesive Targets", accepted for publication at the 64th IEEE Conference of Decision and Control, 2025. 

• I. Napolitano, S. Covone, A. Lama, F. De Lellis, M. di Bernardo, "Hierarchical Learning-Based  Control for Multi-Agent Shepherding of Stochastic Autonomous Agents", submitted to IEEE Transaction on Control Systems Technology, 2025.