Event-triggered adaptive control for delayed memristive neural networks with unknown parameters and external disturbances

The synchronisation problem is studied for master-slave memristive neural networks (MNNs) in this paper. For alleviating the burden of communication bandwidth, a novel event-triggered scheme of data transmission is designed in the sensor-to-controller (S-C) channel. To deal with the unknown parameters and disturbances of master-slave MNNs, the adaptive controller is designed with the system states of triggering instants. Different from existing results about event-triggered adaptive control (ETAC) for MNNs, in which the event-triggered mechanism (ETM) is installed in the controller-to-actuator (C-A) channel, the event-triggered scheme in this paper is designed between the sensor and the controller, so the information flow of S-C channel is discontinuous. The adaptive laws can only use discrete-time system states transmitted at triggering instants to update control gains in this paper. By means of the Lyapunov methods, adaptive control theories and event-triggered techniques, sufficient conditions for synchronisation and quasi-synchronisation are obtained. At the same time, the designed ETM can avoid Zeno behaviour theoretically. Finally, the validity of the obtained results is shown by two simulation examples.