Hybrid tuning of a robotic manipulator controller with a database | Научно-инновационный портал СФУ

Hybrid tuning of a robotic manipulator controller with a database

Тип публикации: доклад, тезисы доклада, статья из сборника материалов конференций

Конференция: International Scientific Conference on Applied Physics, Information Technologies and Engineering (APITECH) / 2-nd International Scientific and Practical Conference on Borisov's Readings; Siberian Fed Univ, Polytechn Inst, Krasnoyarsk, RUSSIA; Siberian Fed Univ, Polytechn Inst, Krasnoyarsk, RUSSIA

Год издания: 2019

Идентификатор DOI: 10.1088/1742-6596/1399/4/044100

Аннотация: In this paper a method of adaptation with hybrid tuning of a robotic manipulator control system with PD controller is considered. The adaptations of the controller parameters and compensating signal are carried out online and independent using modified simplex search algorithms. To increase the performance of the adaptive control system, the values obtained during adaptation process are stored in a database. The database stores these values and the system state, at which they were obtained during the training phase, and in the control phase the database returns the values according to the current state of the system. Efficiency of the proposed methods is verified by simulations. The results show advantage of the adaptive control system with the database. © Published under licence by IOP Publishing Ltd.

Ссылки на полный текст

Издание

Журнал: Journal of Physics: Conference Series

Выпуск журнала: Vol. 1399, Is. 4

Номера страниц: 44100

ISSN журнала: 17426588

Издатель: Institute of Physics Publishing

Персоны

  • Galemov R.T. (Siberian Fed Univ, Dept Robot & Tech Cybernet, 79 Svobodny, Krasnoyarsk 660041, Russia)
  • Masalsky G.B. (Siberian Fed Univ, Dept Robot & Tech Cybernet, 79 Svobodny, Krasnoyarsk 660041, Russia)

Информация о публикациях загружается с сайта службы поддержки публикационной активности СФУ. Сообщите, если заметили неточности.

Вы можете отметить интересные фрагменты текста, которые будут доступны по уникальной ссылке в адресной строке браузера.