Distributed Output Feedback Consensus Control for Nonlinear Multiagent Systems Under Output Event-Triggered Communication

This paper focuses on the leader-following consensus control problem for nonlinear multiagent systems (MASs) under output event-triggered communication. A novel distributed discontinuous backstepping control strategy is presented, which utilizes information exclusively from intermittent output instants. First, a distributed adaptive event-triggered mechanism (ETM), along with a continuous-discrete time compensator and observer are designed jointly to compensate for consensus errors and reconstruct the system state, where all parameters designed are flexibly chosen. The triggering sampling instants are asynchronous and aperiodic, eliminating the necessity for continuous neighbor information monitoring. Second, we introduce a dynamic variable into the coordinate transformation to overcome the challenge of intermittent output in backstepping design, which is the key to address input errors caused by triggering mechanisms. Compared with the widely used backstepping method using the first-order filter to address intermittent signals, the proposed scheme does not require additional triggering design for the filter and achieves a full-state consensus in the global sense. Theoretical analysis shows that the system is asymptotically stable, all consensus errors converge to zero, and Zeno behavior is excluded. Finally, simulation examples are presented to demonstrate the effectiveness of the theoretical result. Note to Practitioners—This paper addresses the challenge of reducing communication costs in multi-agent systems (e.g., autonomous vehicles, sensor networks) where frequent data exchange strains network resources. Traditional continuous or periodic communication methods waste bandwidth and energy. Our solution employs output event-triggered communication, enabling agents to transmit only when necessary. By integrating adaptive triggering thresholds with a simplified control design, we eliminate complex filters or continuous monitoring while ensuring full-state consensus (e.g., synchronized positions/speeds). This approach enhances scalability for large networks and resource-limited systems like drone swarms or smart grids. Future extensions could address unknown disturbances as well as consider time-varying and switching topological networks. The method is ideal for applications prioritizing energy efficiency, such as autonomous logistics, environmental monitoring, or industrial automation.