The optimization of a tanker autopilot control system using genetic algorithms

A genetic algorithm (GA) optimization application is presented in this paper. It involves the performance optimization of a navigation system for a nonlinear tanker model. This navigation system is fully autonomous and regulates the heading of the vessel with reference to a desired course made up of waypoints. The system consists of two components. The first is a line of sight (LOS) autopilot, which determines the desired heading of the tanker from positional information. The second is a heading control system that is derived from sliding mode (SM) control theory. In this investigation the GA is used to optimize the parameters for the SM controller so that the performance of the complete system satisfies specific design criteria. The resulting optimized navigation system is evaluated for different operating conditions for the tanker, e.g., different course, different forward velocities.