Modelling and Hybrid Intelligent Control of Submarine Periscope

In this paper, first, the dynamic equations of a submarine periscope will be extracted and verified with real data. These data are acquired from an experimental setup. Then, using a neural network, a hybrid intelligent control method will be applied to control the periscope model. This control scheme is combined of a proportional controller with a feedforward neural network to adapt the controller to the nonlinearities and parameter changing of the plant. The neural network will be trained on-line without any predefined initial weights to cope with the changes in the system parameters. Simulation results will be compared with the conventional PID controller, which reveals good performance of the proposed controller.