On Fuzzy-PID Control Of A Three-DOF Craft Model

The quadrotor is provided by the Canadian company Quanser experimental apparatus, which has broad applications in both military and civilian region. For example, it can be used for the reconnaissance and surveillance in the ground battlefield, or the search and rescue after serious disasters. These years, more and more research communities have shown a great interest in the quadrotor., many research institutions are working in this area of research. Experimental model system of control can be completed in the laboratory to conduct research control law.This thesis is mainly about the attitude control of the three degree of freedom (three DOF) hovering system with quadrotor. The thesis studies fuzzy PID controller simulation model of the aircraft and control. The text bases on three degrees of freedom for the four rotor hovering system characteristics, mainly for the following studies.The first chapter introduces the four main rotor unmanned helicopter research and related technologies, are discussed in this regard the importance of research work, followed by fuzzy control also introduced the development of the status quo at home and abroad.Chapter II studies of a three-DOF four-rotor hovering system characteristics. Introduce the dynamic model, established the state-space equation system and given some of the parameters of the system neglecting the gyroscopic effect and the friction between the internal structures of the system with the analysis of the system's status of forces.Chapter III researches and designs fuzzy PID controller given the characteristics of the system model which the system is a linear model. PID controller for the linear model is the most commonly used controller, but as a result of PID control parameters need adjustment which is not very convenient, so fuzzy PID controller is used in this article. This controller facilitates the adjustment of parameters and methods of solution for defuzzification uses the different method which literature use of the establishment of TS fuzzy model of solution. And simulation verifies its feasibility. Also compare the PID and fuzzy PID simulation maps and results are analyzed. The results show that the controller in the time and the overshoot is obviously better than the PID controller parameters to achieve the self-adjusting to verify the effectiveness of the method.Chapter IV uses of improved genetic algorithm to generate fuzzy rules automatically considering that the expertise in certain circumstances does not apply and expertise are flawed, therefore the use of genetic algorithms to automatically generate fuzzy rules. And the selection of initial population, the selection of fitness function, crossover probability and mutation probability of selection are chosen for the better one. Moreover, with the use of MATLAB, simulation results show the feasibility of the algorithm. And compared with initial genetic algorithm generating fuzzy rules generated by the control curves, the regulation time and the overshoot are obviously superior to the latter to verify its effectiveness.Chapter V summarizes the major work of this paper, meaning of practical application, prospects for further work and future development prospects.