Analysis and control of Quasi holonimic mobile robot

OData support
Supervisor:
Dr. Harmati István
Department of Control Engineering and Information Technology

The analysis and control of a mechatronic system built during the BSc studies has been investigated in the framework of the thesis. The mechatronic system is a special mobile robot (so-called mobile wheeled pendulum) having two actuated wheels and an intermediate body (pendulum). Due to the mechanical structure the intermediate body tends to oscillate during the translation motion of the robot, thus the application of modern control methods is essential in order to stabilize the dynamical system.

In the first part of the paper, we describe the holonomic and nonholonomic mechanical systems traditionally used in robotics and control as well as the quasi-holonomic system which is based on the fundamental mechanical systems. Therefore we determine the nonlinear mathematical model of the mechatronic system and do the relevant mechanical analysis. In the second part of the paper, we design different control methods to the real mechatronic system and give a comparison regarding the control performances. Our goal was to investigate two control tasks, namely the balance control and the anti-sway control of the robot. Regarding the control methods, the classical LQ(G) optimal control and the soft computing based Fuzzy control were tested and implemented.

During the development, state estimator is designed to the non-measurable states, the LQ problem is described and the optimal state feedback matrices are calculated by the help of the mathematical algorithm. Next the fuzzy logic controller is described and cascade fuzzy control schemes are designed to the system. The comparison of the control performances is based on the simulation and implementation results. The control methods are designed to the nominal parameters, and our interest is to analyze if the implemented algorithms are robust against inertial uncertainties and/or external disturbances.

Downloads

Please sign in to download the files of this thesis.