Search

The optimal output synchronization problem of heterogeneous multi‐agent systems (HMASs) with communication delays is studied in this article. In order to compensate for communication delays actively, we employ networked data‐driven predictive control method to predict the agents' outputs. Then the output synchronization issue of HAMSs is transformed into the stabilization issue of the local pseudo systems, and model‐free off‐policy reinforcement learning technique is used to learn optimal control policies of the local systems. Combined predicted output information with optimal control policies, we develop a distributed data‐driven optimal output policy for each follower. Finally, a numerical example is given to show the effectiveness of our results. [ABSTRACT FROM AUTHOR]

This article studies finite-time distributed formation tracking control of a second-order multi-unmanned aerial vehicle (UAV) system. To reduce the burden of the system's computation and communication, a distributed framework is designed in the control of the formation configuration, in which a time-varying reference trajectory is known by the following UAVs as exogenous input that decides the UAVs moving target. Then, time-triggered and event-triggered consensus control protocols are proposed and theoretically proved to achieve the formation flight in finite time. Finally, a numerical simulation is conducted for the formation tracking control of a typical multi-UAV system to validate the effectiveness of the proposed control protocols. [ABSTRACT FROM AUTHOR]

This study is concerned with adaptive model‐based event‐triggered control of an uncertain continuous system with external disturbance. The proposed framework incorporates two important control techniques for reducing communication burden and regulating the states of the system online in control network, that is, adaptive model‐based networked control system and event‐triggered control (ETC). An adaptive model of the plant is capable of generalising the zero‐order hold implementation in traditional ETC schemes, while also providing stability thresholds that are robust to model uncertainties. In the adaptive model‐based controller, the authors present an update law to estimate the parameters of the adaptive model at triggered instant. After revisiting the adaptive model property in the context of event‐triggered communication, an event‐triggered condition is proposed using the Lyapunov technique. The stability condition of the proposed approach does not need explicit knowledge of the plant parameters, but are given only in terms of the parameters of the adaptive model and some bounds in the model uncertainties. In addition, lower bound on transmission periods are provided. Meanwhile, L2 stability with respect to external disturbance is examined. A real‐time simulation example is presented to demonstrate the effectiveness of the theoretical results. [ABSTRACT FROM AUTHOR]