Probabilistic model predictive control for mineral flotation using Gaussian processes

Abstract

A Gaussian process model predictive control strategy for mineral flotation that builds uncertainty quantification directly into the controller. The Gaussian process is updated in real time as new data arrive, and penalising predicted uncertainty in the objective improved both control performance and robustness.

Dr Paulina Quintanilla

Assistant Professor in Process Systems Engineering

Assistant Professor in Process Systems Engineering at UCL, leading the LOOPS research group. Research interests include machine learning, optimisation and control for processes and physical systems.