Authors: Roland Bartók; József Vásárhelyi
Addresses: Institute of Automation and Info-Communication, University of Miskolc, Miskolc, Egyetemváros, Hungary ' Institute of Automation and Info-Communication, University of Miskolc, Miskolc, Egyetemváros, Hungary
Abstract: Complex control systems are difficult to describe with mathematical models due to the complexity of the system. The control of these systems can be solved with a solution that does not require accurate knowledge of the system model. The most common solution in such cases is the use of neural networks. With complex description of rule-based behaviour, complex control system can also be implemented. Using fuzzy logic is an effective way to describe the behaviour of a system. The problems encountered when using classical fuzzy logic are avoided by using a fuzzy interpolation method such as the Fuzzy Interpolation in Vague Environment (FIVE) method. Compared to classic fuzzy logic, FIVE has the advantage of computing acceleration, through algorithm parallelisation, which is important in the case of real-time-calculations and in tuning procedures. The paper presents a flexible parameterisable hardware-accelerator implementation of the FIVE method using Field Programmable Gate Array (FPGA).
Keywords: fuzzy; FPGA; hardware accelerator; etorobotics; robotics; behaviour-based control.
International Journal of Computer Applications in Technology, 2022 Vol.68 No.4, pp.321 - 331
Received: 04 Mar 2021
Accepted: 21 Jul 2021
Published online: 01 Sep 2022 *