Fuzzy Self-correction Human-simulated Intelligent Coordinate Control And Its Application In Planar Inverted Pendulum

In the field of automatic control, inverted pendulum is known as the Jewel in the Crown. The inverted pendulum is a multi-parameter, nonlinear, strong coupling, unstable controlled plant. It can simulate the verticality control and the flight control of the rocket launch, as well as the stability control of a biped robot and some other advanced technology. So, to some extent, once the stabilization of an inverted pendulum is solved, the key control problem of those nonlinear, strong coupling and multi-parameter systems are solved. And the planar inverted pendulum system, sometimes also known as space inverted pendulum system, has the most number of freedom degrees among inverted pendulum plants. Its engineering background is more realistic, more typical, and thus there are more challenges for its stabilization control issue, which is also the reason of this paper.Human-simulated intelligent control(HSIC) is proposed by Chinese scholar, it reflects human being’s control experience and control logics. HSIC has formed its basic theoretical system and is now being used in many engineering fields. HSIC works well in the test of time and practice. The human-simulated coordinate decoupling control in multi-parameter coupling system is one of the most important idea of HSIC, it also works well in engineering projects. But the traditional human-simulated coordinate control may cause discontinuous interference of the controlled variable. This has a bad effect on its adaptability for more engineering application. And this is the key theoretical problem of this paper.This paper firstly analyze the physical structure and properties of planar inverted pendulum system, then a dynamical model of the system is modeled and linearized in the condition that the rod is almost stabilized near its "inverted" state, in this way, the system is decoupled in X and Y axis. While dealing with the discontinuous interference of controlled variable, the fuzzy control method is used to help to design a human-simulated coordinate control algorithm, by this new control strategy named fuzzy self-correction human-simulated coordinate control, the controlled variables can be adjusted online, in real time and dynamically. And the controller for each control loop of the whole system is nine-point controller, which has a wonderful idea and is similar to HSIC. Finally, the proposed control strategy is used to control the planar inverted pendulum system successfully. The simulation experiment in MATLAB and the real time control experiment show that the proposed control strategy is easy and effective, what’s more, it has a good adaptability and robustness.The main work and the novel properties of this paper are shown as follows.Firstly, the research background, research significant and research status of this paper are introduced. Especially, the research history of the planar inverted pendulum and the HSIC are presented in a few tables.Secondly, the dynamical model of the planar inverted pendulum system is modeled and linearized near the rod’s "inverted" state, in this way, the system is decoupled in X-axis and Y axis, and this will be the breakthrough for the design of its control strategy. Other than other modeling method, instead of calculating the deflection angle of the rod as the combination of the deflection angle in X-axis and Y-axis, the deflection angle in X-axis and Y-axis is used as the direct parameter in modeling. In this way, the derivation of the mathematical equation is simplified and is helpful for people to understand the real control process of the planar inverted pendulum system.Thirdly, the HSIC is introduced briefly, as well as the nine-point controller and the similarity between HSIC and nine-point controller. Most importantly, the novel control strategy named fuzzy self-correction human-simulated coordinate control for multi-parameter, strong coupling control system are proposed, this novel control strategy based on HSIC, nine-point controller and fuzzy control method. The novel control strategy also solved the discontinuous interference of the controlled variable.Fourthly, the proposed control strategy mentioned above is used to design a controller for planar inverted pendulum system. The fuzzy self-correction human-simulated coordinate controller for planar inverted pendulum system can adjust the controlled variable online, continually and automatically. What’s more, the designed quantization factor, scale factor and amplification factor contribute the controller to a better control effect. The simulation experiment in MATLAB and real time control experiment show the effectiveness and robustness of the proposed control strategy. In detail, while in MATLAB simulation experiment, the steady range of the rod with the proposed strategy without modeling the plant is about one hundred percent bigger than the LQR method which is based on the accurate modeling of the system, without modeling the design cycle for the controller can be shorter.Finally, it’s the conclusion and prospect.