Abstract
Thanks to phenomena like superposition and tunnelling, quantum computers have shown the capability to significantly reduce the computational complexity of certain classes of non-convex optimization problems, allowing their solution with a substantial speedup with respect to classical computers. In this perspective, the integration of quantum optimization techniques in the automatic control field has the potential to give relevant advancements in solving complex, high-dimensional problems that are computationally intensive for classical methods. In this talk, we explore the application of quantum optimization in two areas: Nonlinear Model Predictive Control (NMPC) and Scheduling. In the context of NMPC, we first introduce a novel polynomial formulation of NMPC, allowing efficient solution of NMPC optimization problems on classical hardware. This formulation is the basis for the development of a quantum NMPC algorithm: the polynomial structure is reformulated as a Quadratic Unconstrained Binary Optimization (QUBO) problem, which is well-suited for execution on quantum annealers, a particular class of quantum computers that are nowadays commercially available. In the context of Scheduling, we show how relevant space mission planning problems with inherent combinatorial complexity can be reformulated as QUBO problems and then solved using quantum annealers. We present some simulation examples, related to trajectory control for automated ground vehicles and satellite scheduling in space missions. In these examples, the developed quantum algorithms are solved using real quantum annealers, and are compared with state-of-the-art algorithms, running on classical computers. In the talk, we also discuss the general advantages, potential, and limitations of quantum computing, in view of its integration in the automatic control field.
Biographical Information
Carlo Novara received the Laurea degree in Physics from Università di Torino (Italy) in 1996 and the PhD degree in Computer and System Engineering from Politecnico di Torino (Italy) in 2002. He held a visiting researcher position at the University of California at Berkeley in 2001 and 2004. He is currently a Full Professor at Politecnico di Torino in the field of Automatic Control. He is the author or co-author of about 200 scientific publications in international journals and conference proceedings. He has been involved in several national and international projects, in collaboration with Italian and European institutions/companies. He is the co-author of several patents in the automotive field. He is a member of the IEEE committee on System Identification and Adaptive Control, of the IFAC committee on Modelling, Identification and Signal Processing, of the IFAC committee on Modeling and Control of Environmental Systems, and a founding member of the IEEE-CSS committee on Medical and Healthcare Systems. His research interests include system identification, filtering/estimation, nonlinear control, predictive control, data-driven methods, set membership methods, nonlinear optimization, quantum computing, and aerospace, automotive and energy applications.