This project is aimed at designing, simulating and constructing a planar, four wheeled Omnibot that is capable of maintaining balance on top of a cylinder by controlling the cylinders’ linear motion. The successful Omnibot is used to demonstrate and teach the effects of active control of unstable systems.
John N. G. Samarasinghe
Design of a omnibot that can accept multiple types of control systems.
2015-06-05
Control Systems, Euler-Lagrange, Gyroscope, MathWorks® MATLAB, PID, Pole Placement, Simulation, Solidworks®
This project is based on feedback control theory of unstable systems. Many mechanical, chemical and electrical systems exist in the environment around has been naturally driven to stability over time from which once was a high energy state. It is difficult to maintain any system in its unstable state for a long period of time. In order to maintain a mechanical system in its unstable state, the system needs to be controlled actively using feedback from its state in real time. Typically a feedback control system consists of a plant or actuator which is being controlled, sensors which monitor the state of the plant and a controller which performs the necessary calculations to obtain a control signal to track the desired behaviour.
The famous Canadian lumberjack past-time activity, Log rolling relies on the principle of active control for an unstable system. The activity involves two lumberjacks standing on a free log in the middle of a large body of water. The main objective of each lumberjack is to eliminate the other by causing him/her to lose balance. This is done by introducing random disturbances to the log by their legs while maintaining their balance. In this scenario the legs behave as the actuators / plant, the sensors are the lumberjacks’ ears and that detect orientation and the controller is the lumberjacks’ brain. This example is one of the motivations for this project.
The demand for active control is increasing with the development of technology. As more and more applications require systems in their unstable state, they are actively controlled. Such an example is given by the other motivation for this project, which is the Segway Personal Transporter (PT). Segway PT’s actively balances itself at an upright position. Tilting or disturbing the Segway PT in the forward or backward direction will automatically move the Segway PT forward or backward in an attempt to maintain upright. This principle is used to control the Segway PT by a person standing on it.
The design of active control of systems for physical systems has dated back to James Watt’s fly-ball governor in 1970 and has since opened led to many technological innovations. Such examples based on the inverted pendulum include one-wheel electric skateboards [3], electrical suspension systems used in cars [4], Maglev trains [5] and other forms of transportation, aircraft altitude control systems and many other modern robotic systems.
The project is aimed at modelling, simulating, building and programming a planar, wheeled Omnibot that is capable of demonstrating active control. To achieve this, the system is made up of a four wheeled robot (Omnibot) and a free cylindrical object. The Omnibot is placed on top of the cylindrical object to obtain a system that has two degrees of freedom. This system is required to maintain balance on top of the cylinder by controlling the linear motion of the cylinder.
The Omnibot requires proprioceptive sensors to detect the orientation and the position of the degrees of freedom. For this purpose the Omnibot will be equipped with an encoder to measure the rotation of the wheel and an accelerometer/gyroscope to measure the tilt of the Omnibot. A DC motor is required to drive the wheels of the system. An Arduino Uno microcontroller unit is used to control the behaviour of the DC motor based on the readings from the sensors.
The system will be modelled and simulated using software. For this purpose the toolboxes available in the MathWorks® MATLAB are used. The Omnibot will also be built and programmed from the beginning in parallel with the system modelling and controller design respectively.
The completed project is used for teaching purposes. It can be used to demonstrate the behaviour or active control by varying the control parameters in the program or by designing new controllers for the system that causes the Omnibot to behave in a different way.
See other projects and posts in the same category as this post
![]() |
Back to Basics: Building a Buggy | |
Going back to basics and building a buggy to explore my interests | ||
Android, Arduino, SP8266, Micro Controller Unit, NodeMCU | 2020-11-21 | Read More.... |
![]() |
Planar wheeled Omnibot for teaching @ The University of Manchester | |
This project is aimed at designing, simulating and constructing a planar, four wheeled Omnibot that is capable of maintaining balance on top of a cylinder by controlling the cylinders’ linear motion. The successful Omnibot is used to demonstrate and teach the effects of active control of unstable systems. | ||
Control Systems, Euler-Lagrange, Gyroscope, MathWorks® MATLAB, PID, Pole Placement, Simulation, Solidworks® | 2015-06-05 | Read More.... |
![]() |
Line Following Robot @ The University of Manchester | |
This line follower was designed as a part of my second year group project while following my course in BEng(Hons) Mechatronics Engineering at the University of Manchester. In this team of 5 my main role was in control system design and navigation strategy. | ||
Control Systems, Line Follower, Buggy, Sensors, PCB Design | 2013-06-08 | Read More.... |
![]() |
Lap Timer | |
Lap timer | ||
buggy, lap timer | 2013-03-21 | Read More.... |