Title :
Co-design of predictive controllers for wireless network control
Author :
Irwin, George W. ; Chen, Jiann-Jong ; McKernan, A. ; Scanlon, William G.
Author_Institution :
Intell. Syst. & Control, Sch. of Electron., Electr. Eng. & Comput. Sci., Queens Univ. Belfast, Belfast, UK
fDate :
2/1/2010 12:00:00 AM
Abstract :
This study presents a co-design approach to wireless network control systems (WNCS). Experimental studies are carried out to identify and analyse the delays inherent in WNCS, with a new inverse Gaussian model for their statistical distribution proposed. Two different packet-based predictive control schemes are presented. One uses dynamic matrix control (DMC), the other is a linear quadratic (LQ) controller incorporating multiple observers. Lyapunov stability is derived using linear matrix inequalities (LMIs) for the multi-observer LQ control whereas Monte-Carlo simulation results on a cart-mounted inverted pendulum are presented to confirm the performance advantages of packet-based DMC. Finally, results from a real-time implementation of the multi-observer LQ controller are also presented.
Keywords :
Gaussian processes; Monte Carlo methods; control system analysis; delays; distributed control; linear matrix inequalities; linear quadratic control; predictive control; stability; statistical distributions; Lyapunov stability; Monte-Carlo simulation; cart mounted inverted pendulum; delay; dynamic matrix control; inverse Gaussian model; linear matrix inequalities; linear quadratic controller; multi-observer LQ control; predictive controller; statistical distribution; wireless network control system codesign;
Journal_Title :
Control Theory & Applications, IET
DOI :
10.1049/iet-cta.2008.0542