Balancing a two-wheeled Segway robot

Abstract

In this thesis, I designed and constructed hardware for a two-wheeled balancing Segway robot. Because the robot could not be balanced based on a control system derived from the original analytical model, additional system dynamics in the form of frictional losses in the motors were incorporated. A SISO PID compensator and a SISO lead-lag compensator were designed to balance the robot based on the new model; both showed acceptable system responses but were subject to high-frequency oscillation. A SISO state feedback controller was also designed, and it was successful in creating stability in simulation and removing the high-frequency oscillation effects. The robot was rebuilt using new parts that better represented its ideal model, and software was created using National Instruments LabVIEW to control the robot.