Balance And Motion Tracking Control Method For A Two-wheeled Self-balancing Electric Vehicle

Two-wheeled self-balancing electric vehicle, which is also known as two-wheeled self-balancing robot, is a kind of underactuated wheeled mobile workbench. It yields wide application prospects in the field of civilian, industry and military, due to its predominant characteristics, such as compact construction, flexible movement, high maneuverability, convenient operation, pollution-free and so on. The inherent instability of the underactuated dynamics makes it be an ideal test platform for control theory experimentation, the balance and movement control is the key of the research. The control strategy developing processes involved in controlling the two-wheeled self-balancing electric vehicle is mainly discussed in this paper, to realize the task of tracking a desired movement while balancing the system.Firstly, the kinematic model of the two-wheeled self-balancing electric vehicle is established according to its motion in the coordinate system, and the dynamic model is estabilished by the Newton-Euler method and Euler-Largrange method. And an improved dynamic model is obtained by eliminating the nonholonomic constraint in the model established by Euler-Largrange method.Secondly, considering that the two-wheeled self-balancing electric vehicle is a typical underactuated mechanical system and sliding mode control is of better robustness and anti-disturbance ability, an adaptive sliding mode control strategy combined with a virtual control is designed based on the zero-dynamics of nonlinear system, to realize its balance and steering motion tracking control. Meanwhile, to deal with the inevitable friction and model uncertainty, a feedback compensation control method based on adaptive laws is presented to improve interference rejection capacity of the system.Thirdly, in view of the underactuated characteristic, a hierarchical sliding-mode control method is proposed. The whole dynamic system is decoupled into two subsystems:the underactuated balance and longitudinal subsystem and the actuated rotational subsystem, the two subsystems are controlled by hierarchical sliding-mode controller and sliding-mode controller respectively.Finally, a double closed-loop control structure for the two-wheeled self-balancing electric vehicle is developed, to complete the tracking task with an arbitrary desired trajectory and guarantee the stability of internal dynamics at the same time. For the inner loop, a virtual control method for the zero-dynamics established by partial feedback linearization is developed to deal with the underactuated problem of the dynamics, while a direct adaptive fuzzy control method is developed to control the actuated dynamics; and the outer loop uses Lyapunov direct method to design posture controller for the kinematic model. The tracking control is achieved as a result of the coordinated control of the posture/force.The stability of the system with the proposed control strategy is testified by the application of the Lyapunov stability theorem. Numerical simulations are carried out to demonstrate the feasibility of the proposed controllers in Matlab/Simulink environment, and analysis the system tracking performance.