Giorgio Picci

Giorgio Picci is a Full Professor in the Department of Information Engineering at the University of Padova, Italy, holding this position since 1980. His academic career includes prior roles as Lecturer (1970-1975) in Statistics, Researcher (1973-1980) at LADSEB-CNR, and Associate Professor (1975-1980) in Electrical Engineering at the same institution. International appointments include Brown University, MIT, University of Kentucky, Arizona State University, Kyoto University, and Washington University.

He received his Dr. Engineering degree (cum laude) in Electronic Engineering from the University of Padova in 1967.

Professor Picci's research pioneered a geometric framework for stochastic systems centered on Markovian Splitting Subspaces, providing probabilistic analogs of state-space concepts. This work revolutionized understanding of minimality in stochastic models and solved noncausal estimation problems. Current research focuses on:

• Subspace Identification: Addressing closed-loop systems and statistical properties of multivariable system identification
• Reciprocal Processes: Modeling finite-support signals (e.g., images) to eliminate border effects of traditional methods
• Mechanical Systems Identification: Using variational integrators for well-conditioned recovery of continuous-time parameters

His publication record (2002-2014) demonstrates deep integration of control theory, linear algebra, and statistical estimation. Key trends include unifying subspace identification algorithms, advancing covariance extension for reciprocal processes, and developing robust methods for ill-conditioned identification problems in mechanical systems.

Professor Picci leads the Padova research unit that developed breakthrough closed-loop subspace identification algorithms. His collaborations span global institutions including the Royal Institute of Technology (Stockholm) and MIT, with ongoing work on recursive identification and statistical properties of subspace estimates.