CASPIAN JOURNAL
MANAGEMENT AND HIGH TECHNOLOGIES
Intellectual control of underwater robot based on artificial neural network
 |
Read | Svischev N. D., Rybakov A. V. Intellectual control of underwater robot based on artificial neural network // Caspian journal : management and high technologies. — 2018. — №4. — pp. 103-119. |
/103-119.jpg)
Svischev N. D. - student, Astrakhan State University, 20a Tatischev St., Astrakhan,, 414056, Russian Federation, svishchev.nikolai@gmail.com
Rybakov A. V. - Cand. Sci. (Physics and Mathematics), Director of the Institute for Research and Solution of Technological Problems, Astrakhan State University, 20a Tatischev St., Astrakhan,, 414056, Russian Federation, rybakov_alex@mail.ru
Recently, underwater robots have been increasingly used to perform work in difficult conditions in the repair of underwater parts of engineering structures, visual inspection of the bases of drilling platforms, ship hulls, underwater pipelines and cables. As a rule, these are semi-Autonomous devices in which the main tasks of automatic control are to control the position and/or speed of the robot. At the same time, the problems of precise positioning of the robot in the conditions of changes in the flow velocities, keeping the robot perpendicular to the inspected plane position at a strictly defined depth, minimizing errors when moving the device along a given trajectory become relevant. The use of neural networks for the construction of adaptive controllers allows to solve the formulated problems. The problem of positioning in a changing environment can be solved using a neural network controller. The article presents a description of the design, kinematics and dynamics of the robot to perform visual inspections of underwater structures, developed its mathematical model based on the theory of control of underwater vehicles, and the results of simulation in Matlab. The results obtained using the neural network controller model are compared with the results for one of the most common models - pid controller.
Key words: нейронная сеть, подводный робот, интеллектуальное управление, позиционирование, динамика подводных аппаратов, нейросетевой регулятор, корректирующий фильтр, кинематика вязкой среды, neural network, underwater robot, intelligent control, positioning, dynam