FPGA implementation of nonlinear model predictive control

The online computational burden associated with the solution of the NMPC optimization problem is a key issue when applying nonlinear model predictive control (MPC) to fast dynamic systems. In order to improve the computation efficiency, a novel hardware implementation method for NMPC on a field programmable gate array (FPGA) chip is proposed. The optimization problem formed by NMPC is solved by particle swarm optimization (PSO) algorithm. The process speed of NMPC can be accelerated by exploring the parallelism-loop unrolling and pipelining architecture. Real-time simulation test of engine idle speed control (ISC) is performed to verify the NMPC controller. The results show that the designed NMPC-FPGA controller can achieve satisfactory performance for ISC.