Two Self-balancing Robot System Design And Implementation

Two-wheeled self-balancing robot as an extension of a single-stage inverted pendulum with higher-order, unstable, multivariable, nonlinear and strong coupling properties, makes it an ideal verification platform for a variety of motion control algorithm. Meanwhile, the two-wheeled self-balancing robot has characteristics of flexible mechanical structure and variable motion control, making it have a wide range of applications in the changing work environment and the task-complexity situations. For example, it can be used for the space exploration, the battlefield detection and demining in the military field, and short-distance transportation in the daily life. The research and design of two-wheeled self-balancing robot has great value for the development of experimental robotics disciplines and related scientific research.The two-wheeled self-balancing robot presented in this paper uses two DC motor as a power source and establishes of a linear model of a DC motor. This paper uses the Newtonian mechanics to analyze two-wheeled self-balancing robot motion law and establishes a kinematic model. By analyzing energy relations of the two-wheeled self-balancing robot in the process of movement, this paper uses the Euler-Lagrange equation method to build kinetic model. With the Integrated use of the kinematic and the dynamic models, this paper gets two-wheeled self-balancing robot state space representation which provides a theoretical basis for the research of the two-wheeled self-balancing robot control method.Two-wheeled self-balancing robot posture detection accuracy will directly affect the stability of the follow-up control system. In order to improve the control precision of the robot, analyze the characteristics of the current inertial sensor error, this paper establishes an accelerometer and gyroscope data fusion model and uses complementary filter and Kalman filter to fuse robot posture information respectively. The experimental results show that the complementary filt er has properties of simple structure, small amount of computation, but zero drift. Kalman filter has a strong inhibitory effect on zero drift of gyro.For the state equation instability of the two-wheeled self-balancing robot, this paper designs the traditional PID controller and the LQR controller. Using Matlab/Simulink simulation to analyze the two controllers, the results show that the PID controller has a significant effect in a wide dynamic range, but shake easily in the equilibrium position; while the dynamic effect of LQR controller is not obvious, but remarkable in the equilibrium position.Finally, the establishment of two-wheeled self-balancing robot experiment platform is finished. This paper achieves static experiments, dynamic stability tests and dynamic following experiments of the two-wheeled self-balancing robot, and verify the validity of the new algorithm proposed.