Discrete-Time Adaptive Iterative Learning Control by Ronghu Chi & Na Lin & Huimin Zhang & Ruikun Zhang

Author:Ronghu Chi & Na Lin & Huimin Zhang & Ruikun Zhang
Language: eng
Format: epub
ISBN: 9789811904646
Publisher: Springer Singapore

(2)College of Mathematics and Physics, Qingdao University of Science and Technology, Qingdao, China

Ronghu Chi (Corresponding author)

Email: [email protected]

Na Lin

Email: [email protected]

Huimin Zhang

Email: [email protected]

Ruikun Zhang

Email: [email protected]

6.1 Introduction

Distributed control that aims for consensus tasks of multi-agent systems has progressed rapidly with a wide range of applications (Olfati-Saber et al. 2007; Oh et al. 2015; Poveda et al. 2019). Moreover, consensus methods (Jadbabaie et al. 2003; Ren and Beard 2005; Cao et al. 2008; Zhang et al. 2012; Rezaee and Abdollahi 2019; Hua et al. 2020) have attracted numerous attentions as the fundamental and popular problems of distributed control. However, most of these results focus on the protocol design with asymptotic convergence for output consensus when the operation time approaches infinity. In contrast, many systems operate repetitively over a finite duration. For example, the heating, ventilation and air conditioning (HVAC) system of a commercial building works repetitively day after day, or week after week, where the working hours, room temperature, and human activities are all repeatable to a certain degree. For such a finite-time consensus, transient response is most important to ensure a good consensus performance.

Due to the learning ability from control experience, ILC is a excellent method to improve transient response over finite duration. Therefore, many works have been developed in combining consensus control and ILC algorithms. For example, some adaptive ILC consensus methods have been proposed for MASs, where the linear parametric systems (Shen and Xu 2017; Yang et al. 2015; Li and Li 2013), higher order systems (Li and Li 2014; Shen and Xu 2018; Jin 2016), nonlinear parametric systems (Wu and Li 2019; Wu et al. 2018), and nonlinear affine systems (Shen et al. 2019; Li and Li 2014, 2015) are all considered. However, there still exist some challenging problems that impede the extensive applications of these AILC consensus methods.

First, most of these AILC consensus methods (Shen and Xu 2017; Yang et al. 2015; Li and Li 2013, 2014; Shen and Xu 2018; Shen et al. 2019; Li and Li 2015) require identical initial condition as that in the traditional adaptive ILC methods. Second, all the AILC consensus protocols (Shen and Xu 2017; Yang et al. 2015; Li and Li 2013, 2014; Shen and Xu 2018; Jin 2016; Wu and Li 2019; Wu et al. 2018; Shen et al. 2019; Li and Li 2014, 2015) depend on a fundamental requirement that the reference trajectories (or virtual leaders) be strictly repetitive, which greatly limits their practical applications because the desired trajectory often varies with different operations. Continuing the HVAC system of a building as an example, the expected setting temperature of the room may vary every day owing to occupant’s preferences. Third, it is worth mentioning that the AILC consensus protocols (Shen and Xu 2017; Yang et al. 2015; Li and Li 2013, 2014; Shen and Xu 2018; Jin 2016; Wu and Li 2019; Wu et al. 2018; Shen et al. 2019; Li and Li 2014, 2015; Li et al. 2015; Chen et al. 2019) all aim at continuous-time MASs. No related researches have been conducted for discrete-time systems due to fewer mathematical tools available.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.