Loading...

Design and Implementation of Intelligent Memory Control for Flexible Magnetic Robot

Jamshidian, Mohammad | 2023

60 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 56561 (08)
  4. University: Sharif University of Technology
  5. Department: Mechanical Engineering
  6. Advisor(s): Arghavani Hadi, Jamal; Zohoor, Hassan; Nejat Pishkenari, Hossein
  7. Abstract:
  8. The flexible magnetic robot is used for minimally invasive surgeries where there is a complex environment. Precise control of the position of the end of the robot and quick adaptation of the robot to uncertainties are among the most important challenges in this field, where the controller is responsible for compensating the error and controlling the position of the end of the robot. Now, is it possible to create a kind of memory in the system that when faced with the same or similar errors in the same situations or close to previous errors, the system uses its past and compensates the error? The importance of this work is that the response speed of the system is increased and the system can adapt to the environment in less time. In this research, a kind of intelligent controller is designed from the combination of network and fuzzy logic. In this controller, first, the error compensates by a neural network controller. Then the control command used for this compensation is written as a fuzzy rule in the fuzzy rules database. Then, if the system encounters this error or similar errors again, the fuzzy part of the controller tries to reduce the error to a large extent by reasoning from the fuzzy rules, which are actually the same as the system's past, without involving the main controller, i.e., the network controller. By doing this, not only fuzzy rules are generated optimally (fuzzy rules will be generated wherever needed), but also the response speed of the system increases. In order to validate the controller, experiments were conducted using a flexible magnetic robot despite disturbances. As part of the results in the experiments, it was observed that in a rectangular path including 32 points, nearly 19% of the path errors were compensated using the network, and with the rules made from this part, 81% of the path errors were compensated only by using Fuzzy part ("memory"), compensated. It is hoped that the method proposed in this thesis as an adaptive control algorithm can be used in the development of control methods for flexible robots
  9. Keywords:
  10. Intelligent Controller ; Fuzzy Controller ; Neural Network Controller ; Flexible Robot ; Soft Magnetic Properties ; Flexible Magnetic Robot

 Digital Object List

 Bookmark

No TOC