Based On Cloud Model Theory Of Intelligent Control Method For Inverted Pendulum

Intelligent control is a kind of humanoid intelligent behavior characteristics of some of the control strategy, aiming at the complex model parameters, model structure that is difficult to mathematically precise description of the controlled object. The cloud model theory proposed by Li Deyi Academician is based on the idea of uncertainty artificial intelligence, the theory focuses on the uncertainty analysis, combines fuzziness and randomness together, so that uncertainty conversion of qualitative concept and quantitative numerical value is achieved, the method based on this theory has been successfully applied in the stability of the inverted pendulum control, knowledge extraction, data mining, system performance evaluation and so on.As an experimental device, inverted pendulum system plays an important role in the process of teaching and research, owing to its own non-linear, strong coupling, multi-variable and the natural instability, it becomes a method to test a variety of control theory and the ideal model. This paper takes cloud theory as a cornerstone to research stability control problem of single inverted pendulum.First, Newton's mechanics is adopt to establish inverted pendulum system's mathematical model, qualitative control of inverted pendulum mechanism is analyzed, and all kinds of inverted pendulum system's dynamic balance modes are given.Secondly, the basic theory of cloud model is elaborated in this paper, focusing on forward cloud generator and reverse cloud generator of normal cloud model, and cloud deduction of qualitative rules is introduced.Thirdly, this paper is based on normal distribution of the "3En principle" improving membership cloud concept's determination algorithm, the improved algorithm reduces the time complexity of fuzzy concept determination, and it can avoid activating invalid rule during cloud deduction. In addition, this paper takes MATLAB language as a tool achieve forward cloud generator and reverse cloud generator's software implementation, and provide software support for achieving inverted pendulum cloud controller design.Finally, this paper utilizes the improved cloud model control algorithm to control inverted pendulum, and compared with linear quadratic optimal control (LQR) control method, the simulation results show that inverted pendulum system controlled by this method has good dynamic response performance and robustness, and cloud model theory's advantages in dealing with uncertain problems are reflected.