Underwater robot intelligent control based on multilayer neural network

Author

Oskin, Dmitry A. ; Dyda, Alexander A.

Author_Institution

Dept. of Autom. & Inf. Syst., Maritime State Univ., Vladivostok, Russia

Volume

02

fYear

2013

fDate

12-14 Sept. 2013

Firstpage

921

Lastpage

924

Abstract

The paper is devoted to the design of the intelligent neural network-based control systems for underwater robot. New algorithm for intelligent controller learning is derived with usage of speed gradient method. Proposed systems provide the robot dynamics close to reference one. Simulation results of control system for underwater robot dynamics with parameter and partial structural uncertainties have confirmed the effectiveness of approach developed.

Keywords

control system synthesis; marine engineering; multilayer perceptrons; neurocontrollers; robot dynamics; velocity control; control system design; intelligent controller learning; intelligent neural network-based control systems; multilayer neural network; parameter uncertainties; partial structural uncertainties; robot dynamics; speed gradient method; underwater robot intelligent control; Artificial neural networks; Mathematical model; Robots; Trajectory; Vectors; control; multilayer neural network speed gradient method; uncertain dynamics; underwater robot;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Data Acquisition and Advanced Computing Systems (IDAACS), 2013 IEEE 7th International Conference on

Conference_Location

Berlin

Print_ISBN

978-1-4799-1426-5

Type

conf

DOI

10.1109/IDAACS.2013.6663061

Filename

6663061