This paper considers the development of three autopilots for controlling the yaw responses
of an autonomous underwater vehicle model. The autopilot designs are based on the
adaptive network-based fuzzy inference system (ANFIS), a simulated, annealing-tuned
control algorithm and a traditional proportional-derivative controller. In addition, each
autopilot is integrated with a line-of-sight (LOS) guidance system to test its effectiveness in
steering round a series of waypoints with and without the presence of sea current
disturbance. Simulation results are presented that show the overall superiority of the ANFIS
approach.
