Neural network identification of underwater vehicle with hybrid learning algorithm

A new dynamic neural network was constructed by borrowing ideas from Jordan and Elman neural networks. To accelerate the rate of convergence and avoid getting into local extremum, a hybrid learning algorithm by Genetic algorithm (GA) and error back propagation algorithm (BP) was used to tune the weight values of the network. Finally, the improved neural network was utilized to identify the AUV hydrodynamic model. The simulation results show that the new network can remember the history state of hidden layer and tune the effect of the past signal to the current value real-timely. And in the presented network, the feedback of output layer nodes is increased to enhance the ability of handling signals. The neural network by hybrid learning algorithm improves the learning rapidity of convergence and identification precision.