| During the period of 1930s to 1940s, Automatic Control Theory preliminarily formed based on Nyquist, Bode and Wiener's works. The relatively complete theory of frequency method and root locus method for the design of automatic control system was set up through the efforts of numerous scholars after world war II. With the rapid development of modern scientific technology, control object, controller and control tasks&goal are becoming more and more complicated. The classical control theory based on precise modeling and numerical calculation is facing many new problems such as the rapid increase of computing complexity, the more and more difficulty of precise modeling, multi-input &diversification of input information and nonlinear system control. Therefore the modern control theory based on state space method was set up. Compared to classical control theory, modern control theory has been used more widely. It can not only adapt to the system that is single variable, linear, steady and continuous but also to the system that is multiple variable, nonlinear, time-varying, discrete. With the rapid development of scientific technology, the requirement of industrial process control is becoming higher and higher. It focuses much more on the robustness, real-time performance, faulty-tolerance capacity, self-adaptability of the control parameter than the precision of control. Also, with the controlled industrial process being more and more complicated plus the industrial process of typical linearity and uncertainty, many systems cannot be accurately described by mathematic model and modern control method is facing unprecedented challenge. To solve the above problems in modern control system, Large Scale System Theory, Complex System and Intelligence Control are the research directions for the current control theory.With the rapid development of computer science and intelligence information processing in recent years, intelligence control theory and intelligence system is developing very fast too. Among them the representative theories are Expert System, Fuzzy Logic Control, Neural Network Control, Genetic Algorithm and Chaos Control etc. Fuzzy Logic Control is one of the important method of intelligence control is growing very fast in recent years. The substance of Fuzzy Logic Control is nonlinear control. It adopts fuzzy language description system which can not only describe the application system's quantitative model but also can describe its qualitative model. Therefore fuzzy logic can be adapted to any complicated object control. This article describes the advantage of self-organizing fuzzy control through the study of the development and application of fuzzy control theory and the comparison of PID with other fuzzy control algorithm.This article use MATLAB to simulate the algorithm of Self-organizing Fuzzy Control to control the 2-DoF robot arm and illustrate the feasibility and availability of algorithm of Self-organizing Fuzzy Control. To evaluate the feasibility of self-organizing fuzzy controller, I make one set of 2-DoF manipulator control system. The experimental result shows self-organizing fuzzy controller has better control performance than traditional fuzzy controller in reducing errors in the tracing of space trajectory of robot arm joint. |
PDF Full Text Request