Shipman

William John Shipman is researcher at MINTEK, working on machine learning applications within minerals processing and metallurgical plants. His main focus is on reinforcement learning at present. He have published a paper detailing some of this work in tuning proportional integral controllers. He have worked on an embedded model predictive control system for regulating cyanide dosing in a gold leaching circuit, developing the communications software, user interface, data collection and supporting software for the control system library (MINTEK proprietary library). His PhD in Engineering focused on analysing 3D X-ray micro-tomography images of ore samples to extract quantitative measurements. A large component of this work focused on applying machine learning to classifying voxels into mineral classes. His masters degree focused on modelling and controlling a miniature helicopter UAV. The model was developed from first principles and a number of experiments were designed to estimate the parameters of the model. Finally, a non-linear model predictive control system was developed in MATLAB and tested against a simulation of the helicopter UAV.