Generalized Consensus of Discrete-Time Multi-Agent Systems with Directed Topology and Communication Delay
- PDF / 375,896 Bytes
- 11 Pages / 595.276 x 841.89 pts (A4) Page_size
- 117 Downloads / 204 Views
Generalized Consensus of Discrete-Time Multi-Agent Systems with Directed Topology and Communication Delay∗ LIU Huwei · CHEN Xin · GUO Liuxiao · HU Aihua
DOI: 10.1007/s11424-020-9003-5 Received: 3 January 2019 / Revised: 21 May 2019 c The Editorial Office of JSSC & Springer-Verlag GmbH Germany 2020 Abstract Compared with traditional consensus, this paper studies the generalized consensus problem for discrete-time multi-agent systems with directed topology and communication delay. Novel distributed consensus protocols with and without communication delay are designed. Based on the analysis of error dynamical system and graph theory, the generalized consensus is globally asymptotically achieved under suitable conditions without changing the zero row-sums property of Laplacian matrix in networks. Moreover, the sufficient conditions for generalized consensus of communication delay are obtained under directed connections. Finally, some simulations have been provided to verify the theoretical results. Keywords Communication delay, connection topology, discrete-time multi-agent systems, generalized consensus.
1
Introduction
In recent years, the dynamic multi-agents distributed coordination problem has attracted the attention of researchers from diverse areas, which is partly due to its broad applications in distributed optimization of multiple mobile robotic systems, synchronization of coupled oscillators, satellite formation, the directions of fish and birds, and fast consensus in small-worlds. About multi-agent systems, one crucial problem is to put forward control strategies in view of local information enabling the whole system to have a coordinated behavior in communication times, which is called the consensus or agreement problem. Consensus problems have a history of several decades in the field of computer science[1] and are widely used to model various phenomena involving information flow, including flocking[2] and robot formation[3] . Many significant latest results on multi-agent systems have been reported in [4–6]. LIU Huwei · CHEN Xin · GUO Liuxiao (Corresponding author) · HU Aihua School of Science, Jiangnan University, Wuxi 214122, China. Email: [email protected]; [email protected]; [email protected]; [email protected]. ∗ This research was supported by the National Science Foundation of China under Grant No. 61772013 and the Natural Science Foundation of Jiangsu Province under Grant No. BK20181342. This paper was recommended for publication by Editor LIU Guoping.
2
LIU HUWEI, et al.
A great number of significant research theories involving consensus problems of discrete-time multi-agents systems have been proposed. The classical consensus[7] of discrete-time systems is discussed in detail and a theoretical framework for analysis of consensus algorithms for multiagent networked systems with directed information flow is provided in [8]. [9] studies the cluster consensus in discrete-time networks and [10] puts forward the group consensus protocol. Moreover, [11] proposes a kind of couple-group
Data Loading...
