Observer-based Event-triggered Bipartite Consensus of Linear Multi-agent Systems
- PDF / 843,656 Bytes
- 11 Pages / 594.77 x 793.026 pts Page_size
- 16 Downloads / 235 Views
ISSN:1598-6446 eISSN:2005-4092 http://www.springer.com/12555
Observer-based Event-triggered Bipartite Consensus of Linear Multiagent Systems Jianpeng Zang, Xia Chen*, and Fei Hao Abstract: The event-triggered bipartite consensus problem of linear multi-agent systems with a connected structurally balanced signed graph is considered in this paper. The state observer is set to estimate the actual states of the system. In the system structure, two event-triggered mechanisms are configured for each agent. The one is placed in the output side of the agent and the other in the controller. Thus the establishment of the state observer is based on the discrete output information sampled at the corresponding triggering time instants. And the inter-neighboring transmission of the observed states occurs at the moments when the corresponding triggering condition of the agent is satisfied. The triggering conditions designed do not rely on continuous inter-neighboring communications. With the help of the algebraic graph theory and Lyapunov stability theory, it is proved that the bipartite consensus can be achieved. Furthermore, Zeno behavior can be avoided in both sides. The effectiveness of the presented control strategy is demonstrated by a numerical simulation. Keywords: Bipartite consensus, event-triggered mechanism, multi-agent systems, state observer.
1.
INTRODUCTION
In recent years, the consensus problem of multi-agent systems has become a hot topic in the area of control, due to its wide application prospect in the fields of multirobot systems, intelligent transportation system, and satellite scheduling engineering. While wireless communication network nowadays is widely used, the communication bandwidth and driving resources in multi-agent systems are still not enough. Thus, reducing the frequency of inter-neighboring interactions on the premise of ensuring system performance has become a challenging task. To solve this problem, some scholars have proposed the event-triggered control mechanism [1], which can achieve the control goal with limited communication resources. Under an event-triggered control mechanism, the communication between neighbors is usually executed when the pre-designed triggering condition is satisfied. In particular, the triggering condition is generally designed as the ratio of the measurement error (that represents the difference between the latest sampled states and the current one) with respect to a predefined triggering threshold. Thus, the frequency of communication can be adjusted to save communication resources. In [1–5], event-triggered con-
trol was investigated for the first-order and second-order multi-agent systems. For general linear multi-agent systems, valuable results were presented in [6, 7], where the control input of the system is designed based on broadcasted states of the agent and its neighbors. A state estimation model was introduced in [8–10], in which the control input and the measurement error were designed based on the estimated states. However, the control input of e
Data Loading...
