Multi-Shape Classification Algorithm Development for Pick and Place Robot

Project Title

Multi-Shape Classification Algorithm Development for Pick and Place Robot

Project Area of Specialization Mechatronics EngineeringProject Summary

 These days, industries are growing at a very fast pace, and we are required to have efficient and quick solution to meet the demands of a highly populated country like ours. For this purpose, we are designing a system that can pick and place objects using image processing. Such systems are commonly used in modern manufacturing environments.

 In this project, a camera is used to detect the shape of object under consideration. The image is then processed using MATLAB commands. The processed signal is then sent to microcontroller which is responsible for the functioning of servo motors. The motion of servo motors control the robotic arm.

 The robotic arm will pick the object and place it in its respective place. This project is useful for sorting objects according to their shapes with minimum time consumption and higher accuracy.

Project Objectives

1.Object detection in the present frame (shape and color)

2. Location of the object in the frame

3. design of the robotic arm with base,shoulder and wrist

4.Inputs for servos for movement of the robotic arm to reach the desired location

5.Adjustment of the gripper for picking up the object

6.Calculations for required force for handling the object

7.Placing of Object on the Object shaped bins 

Project Implementation Method

1.Algorithm developed for image processing is done on MATLAB software using image aquisation toolbox.

2. After identification in MATLAB, control signal signal is transfered to the microcontroller by MATLAB using serial communication.

3.Microcontroller generates further signals to servos for the Movement of the Robotic arm

4. Required force calculated will control the gripper movement for picking the object.

5. Placement of the object to the specified place.

Benefits of the Project

1.Can be used as a manipulator in industries reducing the manforce.

2. A useful tool to attain fast speed and better accuracy.

3.Involves machine learning technique:image processing and not required fulltime manpower to control the manipulator

4. Further can be utilied in Quality control departments in the packing industries.

5.This protype can be used for to increase production rates in the industry.

6. in small and large industries for sorting of any kind of items on basis of their shapes.

Technical Details of Final Deliverable

1. Microsoft lifecam cinema 720p HD Webcam is the webcam that we are using in this project for capturing the image. Webcam is connected with MATLAB for processing images.  
2. Image Processing Algorithms for multishaped object detection.
3. Beaglebone black is the microcontroller used for getting the processed signals and sending the commands to the servo motors of the robotic arm.
4. Different types of servos are used for different joints. They are HS-485HB, HS-805BB, HS-755HB, Hitec HS-645MG, Hitec HS-422 for base, shoulder, elbow, wrist and gripper respectively.
5. As far as the robotic arm is concerned, we are using Lynxmotion AL5D robotic arm combo kit it has 4 degrees of freedom.It is capable of advance reverse kinematics. Material used is aluminium and it is SES compatible.

Final Deliverable of the Project HW/SW integrated systemCore Industry ManufacturingOther Industries Others Core Technology RoboticsOther Technologies Artificial Intelligence(AI)Sustainable Development Goals Industry, Innovation and InfrastructureRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	68000
Lynxmotion AL5D 4 DOF robotic arm combo kit	Equipment	1	22000	22000
Servo Motors(HS-485HB, HS-805BB, HS-755HB, Hitec HS-645MG, Hitec HS-4)	Equipment	5	6000	30000
Beaglebone black microcontroller	Equipment	1	10000	10000
Sensors	Equipment	2	3000	6000