AN INTELLIGENT CONTROL-SYSTEM FOR REMOTELY OPERATED VEHICLES

Remotely operated vehicles (ROV's) are an important tool for undersea operations. They are used for various tasks like inspection, repairs, and retrieval. The difficulties in modeling an ROV and its hazardous environment restrict the use of conventional controllers. In this paper the application of a neural network controller for ROV's are described. Three learning algorithms for on-line implementation of a neural net controller are discussed with a critic equation. These control schemes do not require any information about the system dynamics except an estimate of the inertia terms. Selection of the number of layers in the neural network, the number of neurons in the hidden layer, initial weights for the network and the critic coefficient were done based on the results of preliminary tests. The performance of the three learning algorithms were compared by computer simulation. The effectiveness of the neural net controller in handling time-varying parameters and random noise were shown by case study on the ROV system.