All Projects
All Projects
Faculty of Engineering
Computers and systems engineering
Graduation Projects
Head of department:
Prof. Hesham Arafat
Project name: One-eyed Self Learning Robotic Arm
Year: 2012
Supervisors: Dr. Amira Yassin
Dr. Abdelhameed F. Ibrahim
Abstract
Factories and workshops in Egypt lack the intelligence in their
machines and still depend on human, and this has bad effects
either on human, like having bad effects on their health that may
cause death, or on Productivity as problems in production, waste
for materials, and cost for each human worker.
All the previous problems lead to use robots with lower cost and
use artificial intelligence and vision system to avoid
reprogramming for each little change in the task.
The project serves in many places with many applications like
painting, welding, drawing, assembling and so on. Robotic arm
is used instead of human as it is more powerful, more accurate,
saves the materials, efficient in repetitive tasks, reduces the cost,
and that all lead to improve Productivity and quality of
production.
At the 1st step, a picture of the object to be painted is captured
using a camera in a fixed position. Some processing is done on
this picture and then, the picture is given an ID and both the
picture and its ID are stored in a database. After that, a new
process called learning begins. In this process, the object moves
until it reaches a position in front of our robotic arm. A joystick
is then used to control the robotic arm so that it can "paint" the
object totally. The movements of the arm robot are recorded and
stored in a database in the microcontroller memory and they are
given an ID similar to the ID that was given to the picture of the
object. This ends the learning process of our system on the
object.
When another object is produced to our system, a picture of it is
captured and compared with the picture of the objects stored in
the database. Here we have 2 cases:-
The 1st case: When the comparison finds a match between
the objects In this case, the ID of the object in the database
is sent serially to our control system so that it can repeat the
recorded movements stored in the database in the
microcontroller.
The 2nd case: When the comparison results in a no-match
situation (The new object is not in the stored database). In
this case, some processing is done on the picture of the new
object. This processing aims at dividing the picture into
groups (grouping) so that we can reach the required points
we need to fully cover the object in the painting process
.These points are then passed to the Inverse Kinematics
part. Inverse Kinematics here is used to find the
corresponding motor angles that are required to move the
arm robot to each point.
Project name: Human arm robotic simulator
Year: 2012
Supervisors: Dr. Mohammed Sherif
Dr. Abdelhameed F. Ibrahim
Abstract
Human arm robotic simulator project is a simulation system
depends on capturing the human arms movements via sensors
attached to his arms and interpret these movements into usable
data; then use these data in controlling a robotic arm in a way
that simulate the same human movements in more accurate way.
The capturing system depends on using sensor based on
calculating the three dimensions and the angle to locate
accurately the location of the node attached to the joint of the
human arm then sends the location information to the controller
to interpret the result in the real time.
The communication system is responsible for transmitting
information wirelessly from the capturing system -through a
controller to organize the operation- to the robotic arm.
A controller stands on the robotic arm receives the instructions
and interprets it into number of degrees to every motor on the
robotic arm joints and by that simulate the human arm
movements.
The human needs the power of the machines but does not need
to be replaced by it, so we are trying to compose a new system
that combine between the human and the machine to empower
the human not to replace him.
This project can be used at many fields and in many
applications, like war field, medical field, mines sweeping field,
save and rescue field, ROV field.
Project name: Smart wireless armed tank (SWAT)
Year: 2012
Supervisors: Dr. Abdelhameed F. Ibrahim
Abstract
Driving robots around a field of obstacles might look like fun.
But for soldiers in combat, those robots in war clearly hold
advantages from saving the lives of our own soldiers. Military
robots are meant to perform tasks that are simply too dull, dirty
and to operating in dangerous environments such as
mountainside and caves to warrant the risk of human life. It also
entertains the point that robots are “unaffected by the emotions,
adrenaline, and stress that cause soldiers to overreact or
deliberately overstep the Rules of Engagement and commit
atrocities.
Swat is robot can be used in battlefield, special mission and
exploring areas because it has the ability to track and hunt objects
based on image processing techniques.
• It can be controlled from large distant because its wireless
features.
• Because we have a camera on the robot, it has the ability to
track, hunt target and secure areas automatically or
manually as user need.
• In automatic mode the ability to track and hunt targets
based on image processing techniques and predefined data.
• In manual mode you can control robot by joystick or by
voice or by using smart phone (android) new way and so
interactive.
• Using a mobile device supports flexibility & the availability
to control & monitor the robot in various places not
restricted to a fixed one.
• Using android as a platform is a plus as android is an open
source OS with less restrictions for developers.
To make it easy in use and control, we developed GUI application
using C# for desktop application or by using android for smart
phone. We used our high experience to make it easy for user to
control the robot either in manual (using joystick or voice or
mobile) or automatic mode (using image processing
techniques).In all GUI modes we have window to preview live
stream from camera.
So, the ides is how to use a small space of area usefully to make
smart park system that park cars automatically in several units.