Neural Network Control of Vehicles: Modeling and Simulation

In the past few years, considerable interest has been shown and relevant resources have been devoted to the design, development and operation of autonomous aerial, underwater, and sea surface vehicles. The possibility of removing human pilots from danger and the size and cost advantages of autonomous vehicles are indeed attractive, but often have to be compared with the performance that can be attained by human-piloted vehicles, in terms of mission capabilities, efficiency and flexibility. The operation of an autonomous vehicle in an unknown, dynamic and potentially hostile environment is a very complex problem, especially when the autonomous vehicle is required to use its full maneuvering capabilities and to react in real time to changes in the operational environment. A common way of dealing with highly complex systems is via a hierarchical decomposition of the activities to be performed by the autonomous vehicles. However, only limited results can be obtained with this method. Another method is to design a hybrid control system that offers safety and performance guarantees by use of neural control technique. Neural networks appear to offer a new, promising direction toward better understanding of the most difficult control problems that have previously been very difficult or impossible to solve. This book provides basic approaches for the modeling and simulation of neural control systems using the MATLAB/Simulink environment for various types of vehicles, emphasizing realistic dynamics with numerous examples. These types of vehicles include experimental aircraft, self-guided missiles, unmanned miniature helicopters, autonomous underwater vehicles, and more.