Page 42 - IJEEE-2023-Vol19-ISSUE-1
P. 42
Received: 20 September 2022 Revised: 26 October 2022 Accepted: 26 October 2022
DOI: 10.37917/ijeee.19.1.5
Vol. 19| Issue 1| June 2023
Ð Open Access
Iraqi Journal for Electrical and Electronic Engineering
Original Article
Robust Control Design for Two-Wheel Self-Balanced
Mobile Robot
Hasanain H. Mohsin *1, Ammar A. Aldair 2, Walid A. Al-Hussaibi 1
1Department of Electrical Engineering Techniques, BETC, Southern Technical University, Basrah, Iraq
2Electrical Engineering Department, University of Basrah, Basrah, Iraq
Correspondence
*Hasanain H. Mohsin
Department of Electrical Engineering Techniques,
BETC, Southern Technical University, Basrah, Iraq
Email: h.h.mohsin@fgs.stu.edu.iq
hasaneen.hadi@gmail.com
Abstract
As a key type of mobile robot, the two-wheel mobile robot has been developed rapidly for varied domestic, health, and industrial
applications due to human-like movement and balancing characteristics based on the inverted pendulum theory. This paper
presents a developed Two-Wheel Self-Balanced Robot (TWSBR) model under road disturbance effects and simulated using
MATLAB Simscape Multibody. The considered physical-mechanical structure of the proposed TWSBS is connected with a
Simulink controller scheme by employing physical signal converters to describe the system dynamics efficiently. Through the
Simscape environment, the TWSBR motion is visualized and effectively analyzed without the need for complicated analysis of
the associated mathematical model. Besides, 3D visualization of real-time behavior for the implemented TWSBR plant model is
displayed by Simulink Mechanics Explorer. Robot balancing and stability are achieved by utilizing Proportional Integral
Derivative (PID) and Linear Quadratic Regulator (LQR) controllers' approaches considering specific control targets. A
comparative study and evaluation of both controllers are conducted to verify the robustness and road disturbance rejection. The
realized performance and robustness of developed controllers are observed by varying object-carrying loaded up on mechanical
structure layers during robot motion. In particular, the objective weight is loaded on the robot layers (top, middle, and bottom)
during disturbance situations. The achieved findings may have the potential to extend the deployment of using TWSBRs in the
varied important application.
KEYWORDS: Two-Wheel Self-Balancing Robot, Simscape Multibody, PID, LQR, Robust controller.
I. INTRODUCTION body (i.e. keeping the robot stable on the horizontal ground)
include proportional integral derivative (PID) [7], Linear
Robotics has been developed and employed widely with Quadratic Regulator (LQR), Linear Quadratic Gaussian control
advanced technology to handle complex designs and activities (LQG) [8], Neural Network control (NN) [9], and fuzzy logic
similar to humans[1]. The use of two instead of four wheels on controlling [10], [11]. Furthermore, three mathematical
robots provides a wide-range of advantages. This includes low methods were used to analyze the dynamical model of
cost, small size, and the ability to rotate quickly (to change the TWSBRs including, Newtonian method [12], Euler-Lagrange
direction and pass through small gaps) due to the need of only method [13], and Kane's method [14]. Nevertheless, an
two points of contact with the floor [2]. Therefore, two-wheel accurate model is required for the design of efficient
robots are utilized in different life-sectors and applications such controllers. It was observed that using the mathematical
as automobiles, rockets, military transportation, public health, equations to model a TWSBR plant had some drawbacks such
industry, and human transportation (Segway) [3]. In particular, as the neglecting of some term for approximate solutions. This
the hardware is used to develop an object-carrying vehicle that has direct impact on reducing the modeling accuracy as well as
can reduce human efforts in the workplace significantly [4]. the dynamical equations must be manually derived and
The Two-Wheel Self-Balancing Robot (TWSBR) can stand in prepared for usage in the block diagram. Basically, it is so hard
the upright position with the help of an inverted pendulum to understand how the physical components of this system
controller. Inverted pendulum stability is well known as a key communicate in practice. Besides, the process of deriving the
issue for dynamically unstable TWSBR [5], [6]. However, essential mathematical equations is rather complex [15], [16].
these works have addressed stability problem for TWSBR To mitigate this issue, the Simscape toolbox in Matlab can be
motion on the simple scenario of straight surfaces. The existing used as an alternative method to model the TWSBR's structural
utilized methods to control the equilibrium position of the robot
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and
reproduction in any medium, provided the original work is properly cited.
© 2022 The Authors. Published by Iraqi Journal for Electrical and Electronic Engineering by College of Engineering, University of Basrah.
https://doi.org/10.37917/ijeee.19.1.5 https://www.ijeee.edu.iq 38