Navigation Control of an Automated Guided Underwater Robot using Neural Network Technique

Abstract

In recent years, under water robots play an important role in various under water operations. There is an increase in research in this area because of the application of autonomous underwater robots in several issues like exploring under water environment and resource, doing scientific and military tasks under water. We need good maneuvering capabilities and a well precision for moving in a specified track in these applications. However, control of these under water bots become very difficult due to the highly non-linear and dynamic characteristics of the underwater world. The logical answer to this problem is the application of non-linear controllers. As neural networks (NNs) are characterized by flexibility and an aptitude for dealing with non-linear problems, they are envisaged to be beneficial when used on underwater robots. In this research our artificial intelligence system is based on neural network model for navigation of an Automated Underwater robot in unpredictable and imprecise environment. Thus the back propagation algorithm has been used for the steering analysis of the underwater robot when it is encountered by a left, right and front as well as top obstacle. After training the neural network the neural network pattern was used in the controller of the underwater robot. The simulation of underwater robot under various obstacle conditions are shown using MATLAB.