Using of artificial intelligence techniques to steering of underwater robotic vehicle along a reference path is considered in the paper. For the tracking, the way-point line of sight scheme is applied. Command signals are generated by an autopilot consisting of four independent controllers with a fuzzy control law implemented. Parameters of the controllers are tuned by means of genetic algorithms. A quality of control is tested without and in presence of environmental disturbances. Selected results of computer simulations are inserted to demonstrate quality of the approach.
This paper considers the decoupled dynamics and control of an Autonomous Underwater Vehicle (AUV). The decoupled model consists of speed, steering and depth subsystems. Generally AUV model is unstable and nonlinear. The central theme of this paper is the development of model predictive control (MPC) for underwater robotic vehicle for ocean survey applications. The proposed MPC for decoupled structure can have simple implementation. Simulation results have been presented which confirm satisfactory performance. Decoupled approach is well suitable for applying control.
Cascaded switching supervisory control of nonlinear underwater robotic vehicle for pipeline tracking
