Scribd
Upload
23 views1 page

Poster

This document summarizes a research project that used a Delta robot with machine vision to classify and sort small products on a conveyor belt. The project involved building a Delta robot model, developing an image processing system using OpenCV to detect and classify products by color, and creating a robot control GUI. Algorithms were developed to control the robot and conveyor belt. Testing verified the robot's path accuracy was within 3% error and positioning accuracy along the z-axis was within 5% error. When sorting 120 products in 5 minutes, the system successfully picked 117 products and accurately classified 117 products, demonstrating stability and minimal misclassification. The project validated the robot's functionality in performing desired trajectories and incorporating color recognition to enhance product identification.

Uploaded by

mtrietdinhbk
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
23 views1 page

Poster

This document summarizes a research project that used a Delta robot with machine vision to classify and sort small products on a conveyor belt. The project involved building a Delta robot model, developing an image processing system using OpenCV to detect and classify products by color, and creating a robot control GUI. Algorithms were developed to control the robot and conveyor belt. Testing verified the robot's path accuracy was within 3% error and positioning accuracy along the z-axis was within 5% error. When sorting 120 products in 5 minutes, the system successfully picked 117 products and accurately classified 117 products, demonstrating stability and minimal misclassification. The project validated the robot's functionality in performing desired trajectories and incorporating color recognition to enhance product identification.

Uploaded by

mtrietdinhbk
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 1

PRODUCT CLASSIFICATION USING DELTA ROBOT

Research Field: Robotics Major: Control Engineering and Automation

1. Introduction
In industrial production, the application of robots in conjunction with machine vision is a trend because it helps increase productivity and reduce the number of laborers.
Furthermore, high-speed Delta robots are particularly suitable for sorting small and lightweight products such as milk cartons, pastries, and candies.

2. Abstraction
Delta Robot model building Control board Image processing

Using image processing algorithm and OpenCV for object detection,


color detection, and camera calibration to find and classify products.

Robot control GUI

Components: Power supply module, MCU Main features: User login/logout, Delta
module, Breakout module, Communication Robot control tools, Robot teaching, Product
module, Stepper buffer module classification module, Product database,
History.

3. Control algorithms and results


Algorithm for Robot Control Algorithm for Control Conveyor

Verification of Robot's Path Accuracy

Procedure: Remove the suction cup of the delta robot and replace it with a ballpoint pen (the length from the pen tip
to the mobile plate should be equal to the length of the suction cup from the mobile plate to the suction cup). Proceed
to have the robot draw straight lines and circles with predetermined diameters (2 cm, 6 cm, and 10 cm).

Conclusion: The system error is less than 3%, indicating that the parameters and kinematic calculations are highly
accurate.

Verification of Robot Accuracy along the Z-axis

Perform the robot's downward movement by 1cm, 2cm, 4cm, 5cm,


6cm, and 10cm. Repeat each value five times.
Comment: With an error margin of less than 5%, it demonstrates
the stability of the robot, indicating that the movement along the Z-
Classification Performance axis is not significantly deviated.

Performing sorting and picking of products on the moving conveyor at a speed of 3 cm/s, the process involves
handling 120 products within 5 minutes. The outcome is the successful picking of 117 products and accurate sorting of
117 products.
Comment: With an error margin of less than 5%, it demonstrates the stability of the robot, as there is minimal
misclassification of products, and only a few are missed.

4. Conclusion
The operational capabilities of the model were verified, and the model was adapted for various project applications. This involved controlling the robot along desired trajectories.
The project also successfully implemented image processing algorithms to identify the color of products, contributing to enhanced product recognition capabilities. Furthermore, a
computer-based robot control program was meticulously developed and achieved successful implementation. Overall, these achievements demonstrate the project's effectiveness
in validating the model's functionality, controlling the robot according to specific trajectories, and incorporating color recognition capabilities into the system, all of which were
facilitated by the development of a robust computer-based control program.

You might also like