Ball-E
Loading...
Date
2025
Authors
Sohaib Arafat
Osama Dweikat
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The project involves a robotic system for ball collection and sorting according to specified
characteristics, such as color, size, or type. For example, tennis balls, ping pong balls,
and beach balls. This robotic mobile arm will be able to move around the playing area,
picking up the balls and sorting them without human interference.
The thing that differentiates our project from the others is the approach. We are trying
to develop a custom Object Detection Model (ODM), for the detection and classification
of the balls, which was trained on almost 30,000 images, supported with an OpenMV H7
or Luxonis OAK-D camera for the best outcome of the job. The arm will be equipped
with various sensors-still to be defined-ensuring retrieval and sorting in an easy way, even
for those balls whose size or type is different. It will be able to move in all directions
with great efficiency due to the use of a stepper and servo motor.
The biggest challenge in this project will be the design of a gripping mechanism that
will be able grasp different kinds of balls without damaging them or the robot rollover.
The gripping arm should be adjusted to variations in ball size and material to ensure a
secure but light grip. The system will also contain sensors for obstacle detection, such
as ultrasonic or infrared sensors, with no interference from movement with the camera.
A control panel gives the user commands for the robot. It will have a Raspberry Pi
main controller for the ODM and camera, while another secondary microcontrollers are
used to control the motors and manage sensors for the robotic arm. This ensures a fully
automated process in the sorting and retrieval of the balls.