Design of a Two-Wheel Self-Balancing Robot with the Implementation of a Novel State Feedback for PID Controller using On-Board State Estimation Algorithm

Abstract

This research was aimed at designing, developing and analyzing a self-balancing robot with the implementation of a novel state feedback for the PID controller using on-board state estimation. Unlike most of the previous works on self-balancing robots, which involved state estimation using IMU and feedback from the actuator side using encoders, potentiometers or tachometers; this research predominantly focused at implementing an on-board state estimation algorithm to generate the feedback, thus eliminating the requirement of sophisticated actuators. Such a system would not only be cost-efficient but would also be more robust in terms of state estimation, since the actuator feedbacks have a recurrently bad impression of misguiding the controller. Subsequently, the robot was designed and a prototype of it was tested for its performance. The performance analysis of the robot exhibited the expected behavior in terms of stability and robustness of the control design.