Model predictive controller design and implementation on FPGA with application to motor servo system

In this paper, an integrated hardware and software design method is developed to implement an MPC algorithm on an FPGA chip. This makes it possible to achieve the long-desired goal of extending MPC algorithms to field control so as to deal with constraints effectively. To expedite the numerical procedure of solving quadratic programming (QP) in the MPC algorithm, a QP solver based on embedded chips is designed to exploit the flexibility and efficiency of FPGA chips. With a carefully devised software architecture, a universal platform is proposed to be facilely deployed to field control applications. To demonstrate the efficacy, a prototype system is built based on a Xilinx FPGA chip. It is then applied to a motor servo tracking control system and achieves satisfactory control performance. (C) 2012 Elsevier Ltd. All rights reserved.