Portal de Periódicos da CAPES

Fast Nonlinear Model Predictive Control on FPGA Using Particle Swarm Optimization

2015; Institute of Electrical and Electronics Engineers; Volume: 63; Issue: 1 Linguagem: Inglês

10.1109/tie.2015.2464171

1557-9948

Fang Xu, Hong Chen, Xun Gong, Qin Mei,

Advanced Control Systems Design

Nonlinear model predictive control (NMPC) requires a repeated online solution of a nonlinear optimal control problem. The computation load remains the main challenge for the real-time practical application of the NMPC technique, particularly for fast systems. This paper presents a fast NMPC algorithm implemented on a field-programmable gate array (FPGA) that employs a particle swarm optimization (PSO) algorithm to handle nonlinear optimization. The FPGA is used to explore the possibilities of parallel architecture for the substantial acceleration of NMPC. PSO is employed to achieve real-time operation due to its naturally parallel capabilities. The proposed FPGA-based NMPC-PSO controller consists of a random-number generator, a fixed-point arithmetic, a PSO solver, and a universal asynchronous receiver/transmitter communication interface. Then, this controller is applied to an engine idle speed control problem and demonstrated with an FPGA-in-the-loop testbench. The experimental results indicate that the NMPC-on-FPGA-chip strategy has good computational performance and achieves satisfactory control performance.