This paper describes the parallel hardware simulation of an adaptive control algorithm. Currently, many modern control systems are designed based on mathematical models. The developed algorithms normally work well in software simulation. However, when implemented into a real-world system, some algorithms appear vulnerable to external disturbance and model imperfection. In this paper, we introduce a hardware-in-the-loop simulation. The simulation test bed, consisting of a Spartan II FPGA board and a PIC 18F452 microcontroller, is coimplemented using both hardware and software and run in a real-time environment. Simulation result demonstrates the effectiveness of the real-time test bed to evaluate an ADALINE based adaptive control algorithm. Further, it can show some problems in physical implementation of the algorithm.
[1]
Martin T. Hagan,et al.
Neural network design
,
1995
.
[2]
Tore Hägglund,et al.
Automatic Tuning and Adaptation for PID Controllers—A Survey
,
1992
.
[3]
Wail Gueaieb,et al.
The hierarchical expert tuning of PID controllers using tools of soft computing
,
2002,
IEEE Trans. Syst. Man Cybern. Part B.
[4]
Fakhri Karray,et al.
Knowledge based approach for online self-tuning of PID-control
,
2001,
Proceedings of the 2001 American Control Conference. (Cat. No.01CH37148).
[5]
Tore Hägglund,et al.
New Tuning Methods for PID Controllers
,
1995
.
[6]
John Wawrzynek,et al.
Reconfigurable computing: what, why, and implications for design automation
,
1999,
DAC '99.