Intelligent Table Tennis Robotic Arm Using Machine Learning and Image Processing

Project Title

Intelligent Table Tennis Robotic Arm Using Machine Learning and Image Processing

Project Area of Specialization RoboticsProject Summary

We are designing a robotic arm that can compete with the humans in Table Tennis with super human intelligencey. The major aspects of this project is image processing, all the work related to image processing will be done in Laptop (i7-7th-gen) and then with the help of USB port we will establish the serial communication between the laptop and Arduino UNO to control the arm. we will use pyfirmata library for the serial communication.

Project Objectives

Our project will follow the four main objectives which are:

Objective #1: Trajectory and position estimation of ball using two camera.

Objective #2: Movement of arm in x and y direction with the help of 1st camera.

Objective # 3: Linear movement of slider depending upon estimated ball position using 2nd camera (z-axis).

Objective # 4: Hit the ball at the particular angle.

Project Implementation Method

1. Trajectory estimation.

To locate the ball in the space we have to use two cameras, one camera will work for XY-axis and the other camera for YZ-axis and we will apply the polynomial regression for trajectory prediction.

2. Get X and Y position with 1st camera.

The X co-ordinate will tell the distance between the arm and the ball, Y co-ordinate will tell the height above the surface of the table. The X co-ordinate will help to hit the ball towards the opponent and Y co-ordinate will adjust the arm bat with the height of the ball.

3. Get y and z position with 2nd camera.

The Z co-ordinate will help the arm to move in left and right direction like a human player, the arm motors will simply follow the direction of the ball in which the ball will going to fall.

4. Send the co-ordinates to the arm connected with Arduino.

The arm should follow the given co-ordinated commands coming through the serial communication from the Jetson Nano.

5. Hit the ball.

Finally the arm has to align with the ball, and on the basis of x-axis we will hit the ball at certain angle towards the human player.

Benefits of the Project

Now-a-days robots are seeking much popularity in terms of training because they are very much accurate, tireless and provides the feedback of the person who is practicing that is the reason people are using machines everywhere for their training purposes. The benefits of the following project are:

1. Now only one player would be able to practice alone.
2. Best trainer for the beginners.
3. Good for physical health.
4. We can play table tennis anywhere and anytime.
5. To practice with the arm is a fun and entertainment as well.
6. 24/7 tireless partner.
7. Provides the Valuable feedback to improve the skill of the player.

Technical Details of Final Deliverable The final deliverable contains are:

• Arduino UNO.
• Two cameras
• Code files.
• Motors.
• Motor Driver.
• Arm Structure.
• Slider for left and right movement.
• Working Demo.
• Documentation of all Components.

Final Deliverable of the Project HW/SW integrated systemCore Industry HealthOther IndustriesCore Technology RoboticsOther TechnologiesSustainable Development Goals Industry, Innovation and InfrastructureRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	48150
Motor Driver	Equipment	3	500	1500
High speed motors	Equipment	5	850	4250
Arm structure	Equipment	1	13000	13000
Motorized slider	Equipment	1	5000	5000
Arduino UNO	Equipment	1	1600	1600
HD Camera	Equipment	2	5500	11000
Table Tennis Bat, printing & travel	Miscellaneous	5	1800	9000
Battery	Equipment	1	2800	2800