ASIC and FPGA Implementation Strategies for Model Predictive Control
Source: University of New York in Prague
Model Predictive Control (MPC) techniques have recently enjoyed an upsurge of interest within the automatic control community, due to their ability to handle non-linear systems and constraints on allowable control inputs and system states. The essential idea underpinning the method is that a constrained optimisation problem is solved in order to generate a required control action. Unfortunately, the requirement for on-line solution of the optimisation problem is a main impediment to widespread implementation of the approach. As such, despite the important attractive features of MPC, its use to date has been largely restricted to chemical process control problems that operate on sufficiently slow time scales.