Learning-enabled event-triggered fuzzy adaptive control of multiagent systems with prescribed performance: A chaos-based privacy-preserving method

Abstract

This paper considers the prescribed performance fuzzy adaptive tracking control problem of multiagent systems under a chaos-based privacy-preserving mechanism and a learning-enabled event-triggered mechanism. Initially, a chaos-based mask function is constructed, which is related to the chaotic states in the Lorentz system. The utilization of chaos adds unpredictability and full randomness to the mask function, which greatly reduces the risk of privacy leakage. Additionally, two value functions are designed as inputs of the fully connected neural network, and the fully connected neural network is used to predict the parameter value in the event-triggered mechanism, which effectively enhances the flexibility of the proposed learning-enabled event-triggered mechanism. Furthermore, in the process of controller design, by employing an error transformed function, the system errors are stabilized within the prescribed performance boundaries. Finally, a simulation example is provided to validate the effectiveness of the proposed control scheme.