Model-Based Distributed Asynchronous Sampling for Multiagent Consensus With Event-Detecting Delays and Network Delays

Abstract

In this paper, the practical average consensus problem of linear multi-agent systems with model-based distributed asynchronous sampling is investigated. Both internal event-detecting delays and external network delays are allowed in systems. Data-sampling instants are generated asynchronously by local event-triggering conditions. Particularly, data-sampling instants and event-triggering instants of each agent are also asynchronous due to the existence of event-detecting delays. Prediction models are utilized in detecting events and updating control inputs. Artificial interpolation instants are introduced in each sampling interval to address theoretical challenges brought by internal and external delays. To deal with this kind of complex asynchronous setup with multiple time delays, based on Barbalat's Lemma, a new procedure for deriving conditions for average consensus and a theoretical analysis approach are presented.