Research On Fuzzy Sliding Mode Control Of Self-balancing Control System

Self-balancing control system is a common control system. Under the control of the computer, the system keeps the angle in the vertical direction or the displacement in the horizontal direction within a certain range through real-time analysis of the system’s states. Therefore the system can maintain its balance. Two-wheeled self-balancing vehicle is a typical underactuated self-balancing system. Compared with other similar systems, its control problem is not only to maintain the pendulum’s static balance in the vicinity of the equilibrium point, but also to maintain its dynamic balance in motion. This thesis selects two-wheeled self-balancing vehicle as the key research object, and designs a new class of fuzzy sliding mode control methods based on sliding mode control to improve the system’s performance in the process of balance control. The research results have certain reference significance for other self-balancing control system. The main research contents are listed as follows:(1) Traditional hierarchical sliding mode control method can be used to control self-balancing system. However, an inappropriate selection of the coefficient of the reaching law will cause chattering phenomenon, which has an effect on the control performance. To solve this problem, this thesis designs a hierarchical sliding mode method based on the idea of backstepping control, and uses fuzzy control to optimize the coefficient of the reaching law. The method improves the control effect on the basis of simple structure of hierarchical sliding mode controller, and suppresses chattering effectively.(2) Using the fixed upper bound is a common method for dealing with the uncertainties in the system. However, it may cause chattering problem. So this thesis uses fuzzy adaptive control to make fuzzy approximation for the uncertainties of the system.In view of the shortage of using a fixed slope of linear sliding mode surface in traditional method, fuzzy method is used to adjust the value of slope in real time to improve the tracking accuracy of the system.(3) Traditional decomposed sliding mode control method can also be used to solve the balance control problem of the system by introducing an intermediate variable. However, this method uses linear sliding mode surface, which is the same as traditional hierarchical sliding mode method. It can not make the system’s states converge to the equilibrium points in finite time. Therefore this thesis designs a fast terminal sliding mode decomposed control method which uses terminal sliding mode surface to replace linear sliding mode surface. At the same time, fuzzy reasoning strategy is used to design a terminal sliding mode surface with time-varying coefficients, which further improves the control performance.