PID Fuzzy and Neural Fuzzy Depth Controllers for an underwater robot: the UROV

Abstract

In this work, a complete strategy of modeling and auto-depth control of an Underwater Remotely Operated Vehicle (UROV) is presented. The effectiveness of the depth control is a position keeping of the UROV such as maintaining the stationary behavior at a depth set point. It is a task to control the attitude of an UROV with a PID, a fuzzy and a neural fuzzy (ANFIS) controllers. The objective of this research is to create an intelligent controller to control the UROV to follow a chosen depth. It is to note that the UROV is progressing in an aggressive milieu filled with a number of disadvantages that are the underwater currents. The UROV must have quick responses with a higher stability and precision facing these turbulences. This technique principle allows some recompenses if compared with the conventional control procedure such as the control with PID. The simulations where done with the Matlab/Simulink program with which a comparative training is completed concerning PID, fuzzy and neural fuzzy controllers beside underwater troubles. A degree of strength is selected for these troubles to examine the strength of the UROV. The results of these simulations are considered to be satisfactory and have definite the effectiveness of the proposed fuzzy and neural fuzzy technics. Effectively, these controllers present better performances in terms of overshoot, settling time and steady state error over PID controller and are efficient for disturbances rejection. Essentially, for the depth control, the system overshoot response should be considered since overshoot is mainly dangerous for the system.