Research Of Adaptive Control Algorithm Of Two-Wheeled Self-Balanced Robot

As a kind of instable wheeled mobile robot, two-wheeled self-balanced robot has multi-variable, non-linear, strong coupling, time-varying and parameter uncertainty characteristics which make it to be an imagine platform to certificate many kinds of control algorithm, so it has very important theoretical significance. Two-wheeled self-balanced robot has flexible movement, simple structure, and it is low cost, suitable for small and dangerous space, so it has a broad prospect.Explore the model reference adaptive control algorithm design-related issues of two-wheeled self-balanced robot, made a foundation for the final realization of any campaign. Through kinematics analysis of robot, we got robot's kinetic energy, potential energy and dissipative energy. Two-wheeled self-balanced robot's dynamic model had been established by using Lagrange equation, and provided a theoretical basis for the controller design. Researched adaptive control method and least squares parameter estimation theory, the model-matching system and recursive least squares parameter identification algorithm was designed. Model reference adaptive controller which based on Diophantine equation was designed through the combine of model-matching system and recursive least squares parameter identification algorithm. Achieved track control of two-wheeled self-balanced robot's title angle, forward speed and turn angle.Adaptive controller's advantage on anti-jamming and adjust time had been proved by simulation for adaptive controller and traditional controller. Through experiments in balance control, anti-jamming and speed tracking the result of simulation had been proved and the model reference adaptive controller's validity had been confirmed.