The thesis presents the construction, modelling and control of an inverted pendulum with a reaction wheel. The first part introduces the pendulum and the possible methods for manipulating its tilt, from which I choose the reaction wheel. This is followed by the derivation of the mathematical description of the system's dynamics using the Lagrange function and Laplace transform, which I then used to determine the transfer function of the system.
The second part presents the realization of the physical model. It describes the process of modelling and 3D printing of the system's components, such as the pendulum, wheel, and base. The operation of the measurement instrument, actuator, electrical circuits, and microcontroller used in the system is also described. This is followed by the assembly of the physical model and the programming of its functionalities, with a more detailed description of the method used to filter the obtained measurements.
The third part covers the simulation of the mathematical model in the Matlab and Simulink environments, where some unknown system parameters are determined using an optimization method. The properties of the derived model are analysed based on the obtained parameters, and the model is then visualized and compared with the real model to validate the accuracy of the mathematical system.
The final part of the thesis delves into control theory, where PID and LQR controllers are presented as widely used methods for system control. Both methods are implemented on the mathematical model, and their results are compared to determine their suitability for the given system.
|