Rapid Prototyping of Constrained Linear Quadratic Optimal Control with Field Programmable Analog Array

Recent advances in embedded model predictive control (MPC) have paved the way for MPC technology in highspeed systems and resource-constrained applications. In this paper, we exploit analog circuit emulation of constrained Linear Quadratic (LQ) problem for rapid prototyping of MPC on fast reconfigurable analog processors such as Field Programmable Analog Arrays (FPAA). The inherent massively parallel structure of the underling solution dynamics and the possibility of signal scaling through appropriate pre-conditioning mean that MPC can be reliably implemented using low-power ultrafast analog circuits. The effectiveness of the approach is demonstrated for a two-integrator MPC problem.