Model-based event-triggered neural learning formation transformation method for multiple unmanned ships

The invention provides a model-based event-triggered neural learning formation transformation method, device and equipment for multiple unmanned ships, and a medium. The method comprises the following steps: constructing a strict feedback system of the multiple unmanned ships; constructing a radial basis function neural network; the adaptive neural network controller based on distance error estimation is used for training a neural network based on historical data to obtain each weight under a convergence condition; acquiring wind direction data based on a wind direction and wind speed sensor carried by each unmanned ship, and determining the magnitude and direction of wind force borne by each unmanned ship; based on a set model event triggering strategy of the neural network model, triggering formation transformation of each unmanned ship system, and feeding back distance error data between a navigator and each follower to the adaptive neural network controller; the technical problem that a neural network mod