Event‐triggered adaptive deep neural optimized containment control for fractional‐order nonlinear multi‐agent system

Abstract In this paper, an event‐triggered adaptive optimized containment control strategy is proposed based on deep neural networks (DNNs) for nonlinear fractional‐order multi‐agent systems. DNNs, with their more complex multi‐layer nested architecture and exceptional function approximation capabilities, are employed to estimate unknown functions. Subsequently, an optimized containment control scheme based on DNNs is presented. Different from the traditional critic‐actor structure, the cost functions are approximated by DNNs, which reduces the control effort. At the same time, the event‐triggered mechanism is used to reduce the consumption of communication resources. The results can ensure that every signal of the system is bounded. Two simulation examples demonstrate the effectiveness of the proposed algorithm.