Eventos Anais de eventos
COBEM 2017
24th ABCM International Congress of Mechanical Engineering
Nonlinear Modeling, Simulation and Control of a Two-Wheeled Inverted Pendulum
Submission Author:
Gustavo Diniz Silva , MG
Co-Authors:
Gustavo Diniz Silva, Bruno Luiz Pereira, Alexandre Sousa, Leonardo Sanches, José Jean Paul Zanlucchi de Souza Tavares, RODRIGO HIROSHI MUROFUSHI, Rodrigo Nogueira Cardoso
Presenter: Bruno Luiz Pereira
doi://10.26678/ABCM.COBEM2017.COB17-0965
Abstract
In the study of nonlinear systems, the modeling process has an important role in the control algorithm’s performance. Once the quality of a model is assured, it can be used effectively in control techniques. Inverted pendulum systems’ control is a classical issue in this field, and so, improvements in those techniques are usually studied. This work aims to improve the performance of PID and fuzzy controllers by adapting them to a nonlinear model and evaluate them through simulation before applying them to the real system. The nonlinear dynamic model of the two-wheeled inverted pendulum is obtained through the Newton-Euler formulation and the linear model is obtained linearizing the nonlinear model. Normally, due to the nonlinearity of the model’s equation, it is impossible to solve them analytically, requiring a numerical approximation. For this purpose, the 4th order Runge-Kutta method was used. In this paper a PID controller using the linear model (LPID) was proposed, and classic PID controller is designed and adapted to the nonlinear model (NLPID). Using a NLPID controller, the simulation and vehicle’s response in real time were compared to statistically validate the nonlinear model. Then a simulation was performed comparing the performance of the controllers, as well as the LPID performance for the linear model and the nonlinear model for different sampling times.
Keywords
Inverted pendulum, Nonlinear System model, Runge-Kutta Method, fuzzy control, PID control
