Sigmoid Surface Controller for Autonomous Underwater Vehicles by Improved Particle Swarm Optimization

Sigmoid surface controller has been proved to be effective in motion control of autonomous underwater vehicles, but it is hard to adjust its control parameters. Improved particle swarm optimization of sigmoid surface controller was proposed in this paper, which solves the problems that particle swarm optimization may be trapped in local optimum and fails to converge to global optimum. Moreover, intelligent integral was introduced to eliminate steady-state errors and increase the stability of sigmoid surface controller. Finally, simulation experiments are carried out on WL-II mini autonomous underwater vehicle. The results show that improved particle swarm optimization converges faster and the optimized sigmoid surface control parameters are more approximate to global optimum, so it makes great improvements to sigmoid surface control in response speed and overshoot.