Research interests
My research develops along two main lines: distributed computing for autonomous and fault-prone systems, and machine learning and big-data methods for digital health.
Distributed computing and autonomous systems
My work in distributed computing began with the design and analysis of algorithms for teams of fully autonomous mobile robots operating in the plane. This research studies how limited local capabilities—such as visibility, memory, synchrony, and communication—determine which global tasks a robot team can solve.
Mobile robots
Distributed coordination algorithms for pattern formation, gathering, flocking, and intruder-related tasks under limited visibility and asynchronous execution.
Simulation and experimentation
Design of Sycamore, a modular 3D simulator for mobile robot systems, and its integration with the VIRCA virtual collaboration platform within the EU VisionAir initiative.
Programmable matter
Local-interaction algorithms for large populations of computational particles, with applications ranging from smart materials to self-repair and minimally invasive technologies.
Fault-tolerant distributed systems
Black-hole search by mobile agents, resilient routing in networks with faulty links, and distributed algorithms for dynamic and intermittently connected networks.
Parallel algorithms
Algorithms for coarse-grained parallel machines and PRAM models, including graph and computational-geometry problems.
Community building
Co-coordination of the Research Meeting and School on Distributed Computing by Mobile Robots, which reached its tenth edition in 2025.
Digital health, AI, and rehabilitation
Since 2017, I have worked on technology transfer and research at the intersection of machine learning, multimodal data analysis, wearable sensing, and healthcare. This activity began with sports analytics and expanded toward clinical applications in cardiology, neurological rehabilitation, cerebral palsy, autism research, endocrinology, and tele-rehabilitation.
Sports analytics and wearable sensing
Machine-learning methods that combine smartwatch and video data to characterize playing style and provide personalized feedback, initially with a focus on tennis.
Clinical decision support
Predictive models based on clinical, laboratory, imaging, and wearable-sensor data, including applications in cardiology, thyroid disorders, and patient stratification.
Personalized rehabilitation
AI-based tools for motor and cognitive rehabilitation, outcome prediction, personalized treatment planning, and continuity between clinical and home-based care.
Multimodal pediatric health
Software support for diagnosis and rehabilitation in children with unilateral cerebral palsy, integrating wearable sensors, magnetic resonance imaging, clinical records, and laboratory data.
Research laboratories
Scientific coordination of the Digital Health area of Acube Lab and co-scientific responsibility for the joint InnoDeep research laboratory.
Education in digital health
Co-promotion of the University of Pisa MSc programme in Informatics for Digital Health, accredited in 2024 and launched in the 2024–2025 academic year.