Abstract:The event-triggered distributed average tracking (ETDAT) problem for heterogeneous multiagent systems (MASs) with uncertain dynamics is investigated in this article. The ETDAT algorithms aim to build control laws for heterogeneous agents to follow the average states of multiple time-varying input signals in event-triggered communication networks. The uncertain dynamics of agents and the event-triggered communication mechanisms make the design of distributed average tracking (DAT) protocols difficult. To achieve ETDAT for heterogeneous MASs with uncertain dynamics, we designed two kinds of ETDAT protocols. First, on the basis of model reference adaptive control (MRAC) technology and sampling measurements, we present a class of static-gain ETDAT algorithms. In comparison to conventional DAT, the proposed ETDAT algorithms not only solve the DAT problem of heterogeneous MASs but also greatly reduce the cost of network communication. Second, dynamic-gain ETDAT algorithms based on self-adaptive principles are presented to minimize network global information needs. The above two algorithms adopt boundary layer approximation methods and dynamic event-triggered strategies, which can further reduce the chattering phenomenon and event-triggered frequency. Finally, the theoretical findings are shown with several examples.

Event-Based Distributed Average Tracking for Second-Order Heterogeneous Multiagent Systems With Uncertain Dynamics