Event-triggered sampling scheme for pinning control in multi-agent networks with general nonlinear dynamics. Issue 8 (November 2016)

• Record Type: Journal Article
• Authors: Li, Huaqing; Chen, Guo; Xiao, Li
• Is Part Of: Neural computing & applications. Volume 27:Issue 8(2016)
• Page Start: 2587
• Page End: 2599
• Publication Date: 2016-11
• Subjects: Event-triggered -- Distributed control -- Pinning controllability -- Complex dynamical networks -- M-matrix; Neural networks (Computer science) -- Periodicals; Neural circuitry -- Periodicals; Artificial intelligence -- Periodicals; Neural Networks (Computer) -- Periodicals; Réseaux neuronaux (Informatique) -- Périodiques; Réseaux nerveux -- Périodiques; Intelligence artificielle -- Périodiques; 006.32
• DOI: 10.1007/s00521-015-2027-4
• Languages: English
• ISSNs: 0941-0643

Abstract

Abstract Event-triggered control strategy of complex dynamical networks is motivated by further applications of embedded microprocessors equipped in the nodes with limited computation and storage resources that gather information and actuate the individual controller updates. One expects that the numbers of controller updates and measurement broadcasts decrease significantly. This paper presents a novel framework of distributed event-triggered asynchronous intermittent communication strategy for pinning controllability in complex dynamical directed networks in which each node is equipped with nonlinear dynamics with or without time delay. The proposed method extends the previous works on pinning control by considering limited communication capacities through strictly noncontinuous decentralized information exchange and considers the existing continuous control schemes as special cases when the corresponding tuning parameters in the designed events are equal to zeros. Based on properties of M-matrices, some analytical criteria are established to guarantee that all follower nodes in a network can globally asymptotically converge to a homogenous state of a leader node. Control inputs exerted on all follower nodes are only triggered at their individual event instants, which reduce the amount of communication in information channels and lower the frequency of controller updates to favor a digital implementation. A positive lower bound for the broadcasting period, i.e., the difference between two successive broadcasting time instants, can be achieved, which excludes the Zeno triggering behavior before the synchronization is achieved. Simulation results show effectiveness of the proposed approach and illustrate correctness of the theoretical results.