Robust distributed model predictive control under actuator saturations and packet dropouts with time-varying probabilities

• Published in: IET control theory & applications Vol. 10; no. 5; pp. 534 - 544
• Main Authors: Han, Huaxiang, Zhang, Xiaohua, Zhang, Weidong
• Format: Journal Article
• Language: English
• Published: The Institution of Engineering and Technology 21.03.2016
• Subjects: actuator saturation; Actuators; augmented distributed controller model; computational effort; Control theory; Controllers; controller‐to‐actuator channel; convex hull; convex optimisation problem; convex programming; Dropouts; Feedback; flexible structures; flexible system structure; global performance function; global system; linear matrix inequalities; linear matrix inequality; linear systems; lower‐dimensional subsystem; Mathematical models; packet dropout; polytopic uncertain system; Predictive control; probability; robust control; robust distributed model predictive control; saturated linear feedback law; Saturation; stochastic systems; stochastic variable; time‐varying probability; uncertain systems; packet dropout; computational effort; global system; feedback; stochastic variable; linear matrix inequalities; stochastic systems; robust control; global performance function; actuator saturation; augmented distributed controller model; saturated linear feedback law; flexible structures; controller-to-actuator channel; uncertain systems; convex optimisation problem; probability; convex programming; linear systems; convex hull; linear matrix inequality; polytopic uncertain system; flexible system structure; robust distributed model predictive control; predictive control; time-varying probability; lower-dimensional subsystem
• ISSN: 1751-8644; 1751-8652
• DOI: 10.1049/iet-cta.2015.0706

In this study, the authors investigate distributed model predictive control for polytopic uncertain systems subject to actuator saturations and packet dropouts with time-varying probabilities in controller-to-actuator channels. For the sake of less computational efforts and a flexible system structure, the global system is decomposed into several lower-dimensional subsystems and the subsystems optimise a global performance function in parallel. A novel augmented distributed controller model is proposed to describe both the actuator saturations and packet dropouts in one unified expression by adopting a stochastic variable taking the value 1 or 0 with time-varying probabilities and accordingly a new result is obtained. By placing the saturated linear feedback law into the convex hull of a group of linear feedback laws, the distributed controller for each subsystem is obtained from the solution of the convex optimisation problem involving linear matrix inequalities. A numerical example is supplied to demonstrate the effectiveness of the developed techniques.