The Application Of Granular Computing Theory And Its New Algorithm In Intelligent Control

Intelligent control is the product of combination of control theory and AI,and is a hot research topic in control field.As a kind of fuzzy control based on natural language control rules,fuzzy inference machine computer control technology,fuzzy control has become one of the most extensive research and application field in the intelligent control technology because of its simple structure,strong robustness,without having to rely on the advantages of accurate mathematical model of controlled object.However,more input variables in the fuzzy controller can produce a large number of fuzzy rules,the system is complicated;In order to simplify the fuzzy control algorithm,currently only able to save to blur steps in the process of fuzzy control,the blur is still a must step.With the great progress of fuzzy mathematics in recent years,especially the emergence of grain computing theory,it provides the necessary foundation for solving this problem thoroughly.A new fuzzy control algorithm based on granular computing theory is proposed in this paper.The new algorithm can simplify the existing fuzzy controller,and design an adaptive fuzzy control system model based on granular function according to the algorithm.The new algorithm is a fuzzy control algorithm which does not use fuzzy logic,The fuzzy input and fuzzy output are granulation,and the fuzzy information particles are obtained.According to the experience and knowledge of the people,we get the particle response function's sample points and fit out the granular function.We use the mapping method to record a mapping relation of granular functions,and fit a function to realize this function.We use this function as the point response function of fuzzy controller to control.This algorithm has the following advantages: first,the point response function can replace the high dimensional fuzzy controller to control,solve the problem of high dimensional fuzzy controller is difficult to avoid the high dimensional modeling,fuzzy control in the control because of too many rules of the system due to the increasing complexity of the problem;second,a new algorithm based on fuzzy control the system eliminates the need for the fuzzy control in the process of fuzzification and defuzzification step,which greatly simplifies the process of fuzzy control;third,the new algorithm based on the fuzzy control system can adapt to theneed of self control precision according to different sampling rate and achieve the function of control of control object.The simulation results of two wheeled self balancing robot control system show that the fuzzy control algorithm based on granular computing theory has a good control effect.The higher the number of the highest response times is,the better the control effect of the system is.