Modeling of self-organization fish school system by neural network system

Abstract EN

In this paper, a Neural Network (NN) model system for self-organization fish school system is identified.

Monitoring and data extraction from fish school video has been achieved by using image processing technique in order to generate the data suitable for parameter identification of NN model system.

Data obtained have been used to identify the parameters of a model based on a black-box represented by nonlinear autoregressive exogenous model (NARX).

The obtained results show that this system can be used for multi robot formation system.