About Me

I am a Lecturer in the Robotics and AI group at University College London (UCL), a position I have held since 2024. Prior to this, I joined UCL in 2022 as a Post-Doc researcher for the RoboHike FLF project, focusing on autonomous navigation for quadruped robots. I received my master’s degree cum laude in Control Engineering at “La Sapienza” University of Rome in 2013.

In 2017, I obtained my Ph.D. from the same university with a thesis on the design of whole-body controllers for humanoid robots, combining control techniques and machine learning. My Ph.D. resulted from a scientific collaboration between La Sapienza and the INRIA Grand-Est research institute under the supervision of Dr. Serena Ivaldi. During my doctoral studies in 2015, I was also a visiting researcher at Darmstadt Technical University with the Intelligent System Team led by Jan Peters.

My research sits at the intersection of modern AI and classical control theory. I specialize in hybrid methodologies, bridging data-driven adaptability with analytical rigor by integrating Model Predictive Control (MPC) and black-box optimization with neural architectures like Vision-Language-Action (VLA) world models. A key focus is establishing formal stability and robustness guarantees for learned policies on highly redundant systems. Building on my foundational work with humanoids and quadrupeds, my application field encompasses underwater, aerial, and extreme robotics for ecological and agricultural tasks. In these unstructured environments, I aim to develop fault-tolerant, "antifragile" autonomous agents that dynamically adapt to unexpected disturbances and sensory failures in the wild.

At UCL, I teach across the MSc in Robotics and AI, the MSc in System Engineering for the IoT, and broader cross-disciplinary MEng and BSc programs. My curriculum bridges machine learning, control, and full-stack embedded systems. My robotics modules focus on advanced autonomy, covering terrain navigation in Legged Robotics, core algorithms to deep networks in Machine Learning for Robotics, and simulation-based Reinforcement Learning (Isaac, MuJoCo) in Robotic Systems. For the IoT programs, I emphasize deployable hardware. I teach end-to-end integration—from board design to AWS/Azure cloud computing—in Modelling and Designing Embedded Systems, and guide students from theory to real-world application in the hands-on, project-driven Designing Sensor Systems module.