An Assessment between Classical Fuzzy and Fuzzy Model Reference Learning Controller Design of Electro-Optical pointing and tracking System

Abstract

ABSTRACTElectro-optical pointing and tracking systems (EOPTS) have a wide range of military andcivilian applications. The passive line of sight (LOS) stabilization systems are multi-inputmulti-output (MIMO) systems that are highly nonlinear and possess a strong coupling effectbetween their states. it presents a challenging systems to control. In this paper the analysis ofthe passive (EOPTS) stabilization system with the development of its nonlinear models isderived. Two different types of control algorithms are presented. The first controller is aclassical fuzzy control. The controller presents a good technique that proves to be stable withhigh transient and tracking performances. The controller is applied to the LOS stabilizationsystem and the simulation results are introduced. Next, A Model Reference Learning fuzzyController less dependant on the designer knowledge of the LOS stabilization system isproposed. Such controller possesses a learning mechanism that is able to form its rule-base bywatching the system behavior. The learning mechanism utilizes a reference model thatdescribes the desired performance. The designed process of the controller and simulationresults of implementing the controller to the system is introduced. Finally, comparativeanalysis between the two developed controllers is conducted. It discusses the advantages anddisadvantages of each controller algorithm.