KFUPM logo


 King Fahd University of Petroleum & Minerals

 Electrical Engineering Department



Capstone Projects

Term:  162 (2017)

Name:  Optical Mouse-Based Odometer and Motion Tracker

YouTube: https://youtu.be/gwXvvmw8Qg0


The project aims at constructing an inexpensive odometer to record, online, the trajectory and orientation of a moving platform. The optical coupling between the platform and its background is carried-out using two standard computer wireless optical mice. Tests were designed and carried-out to determine the usability of the mice as measuring device. The mice outputs are found to suffer from a high amount of noise. Different filters were tested to suppress this noise. A configuration for setting-up the mice in order to use them as an odometer is suggested. A procedure is proposed for processing the mice output in the adopted setting  to produce the position and orientation estimates. An RC platform was also designed and built to carry the odometer and experiment with it. Experiments were designed to calibrate the odometer as a whole system.  Thorough experimentation were carried-out to assess the capabilities of the system.

Term:  152 (2016)

Name:  An Optical Targeting System

YouTube:  https://youtu.be/6wdkDIYCwuY

The project won 3rd Place in the EE design EXPO


The project tackles the design and implementation of a basic optical targeting system. The system is required to detect the entry of an object into the camera visual field, track the object while maintaining an LED pointer lock on the target. It consists of the following subsystems: the mechanical pan-tilt platform along with the needed computer interface, an optical processor and an orienting controller to interface the processor with the pan-tilt platform. The design emphasizes the use of affordable hardware. It also restrict processing to an ordinary laptop and matlab software. Both the hardware and software environments were developed and interfaced. Thorough experimental testing was carried-out.

Term:  151 (2015)

Name:  A Visually-servoed, Radio Controlled Toy Car

YouTube:  https://youtu.be/yysk84lVGkE


A  fully autonomous (push button), visually-servoed system for a car-like robot operating in free space is built from the ground up.  The aim is to produce an inexpensive system with almost all the functionalities needed to operate as a test-bed for developing controllers, trackers and other modules needed for implementing such systems. The car used is an inexpensive RC toy car that cost about 30 Saudi Riyals (9 USD). The workspace of the car is visually monitored using a standard 10-USD PC webcam. The handset of the car was interfaced to matlab via the Aurdino microcontroller (will be removed in the future). Inexpensive Interface circuitry (less than 10 USD) was built so that commands to the car can be sent electronically. Software is developed to online estimate the state of the car from the sequence of images steamed to matlab. Three controllers are developed to use the car state and the target point from the computer mouse  to generate an online control hat will move the car from any where in the visual filed to the specified target point.

Term:  141 (2014)

Name: An Intelligent Optical Maze Solver

YouTube:  http://www.youtube.com/watch?v=W6kxll6aqX4


An intelligent  vision system is constructed that is able to solve any maze a human operator present it with. The maze is fed to the system using a computer webcam. The start and end points are marked using a computer mouse. Edge detection is used to segment the image into obstacles which the path has to avoid and free space which the path can pass trough. Electromagnetism is used to construct the path by treating obstacles as an insulator and free space as a perfect conductor.

Term:  132 (2014)

Name: Maze Solving Using the X80 Mobile Robot

YouTube:  http://www.youtube.com/watch?v=oSCz1NIUwYs


A navigation control is designed for a differential drive robot to move the robot in an unknown environment from one point to another. The navigator on-line senses the environment using ultrasonic sensors. The Harmonic potential approach is used to convert the sensor data into an obstacle-free  trajectory for the robot linking the start and end points.

Hit Counter

Disclaimer: Material are provided on this page for the sole purpose of timely information dissemination, and the copyrights belong to the authors and the corresponding publishers unless otherwise stated.


Copyright © 2008