YAFT: A Fuzzy Logic based Real Time Two-Wheeled Inverted Pendulum Robot

Abstract

This paper presents the mechanical design of a custom built two wheeled inverted pendulum, named as YAFT, and an intelligent control system structure to solve its control problems. YAFT is equipped with various sensors, from IMU to camera, and processing units, from STM to Raspberry PI, that give the opportunity to deploy YAFT in telepresence, cognitive robotic and human-machine applications. As we have preferred to use low-cost market components, we perform a deflection analysis to show that YAFT has a rigid body frame and then derive its dynamical model. We propose also an intelligent control structure that is capable to handle the real-world uncertainties resulted from low-cost components. The proposed intelligent control structure takes up and merges the frameworks control theory, fuzzy logic and sensor fusion in order to solve the stabilization and position control problems of YAFT. Comparative experimental results are presented to illustrate the effectiveness of the proposed control structures on the YAFT in real-time.