Automated Fruit Harvesting Robot

Project Title

Automated Fruit Harvesting Robot

Project Area of Specialization Electrical/Electronic EngineeringProject Summary

Agriculture and labor saving in agriculture is in the need of modern era. However, mechanization for fruit harvesting has not been advanced yet. This project proposes a method of detecting fruits and automated harvesting using a robotic arm by digital image processing.A highly fast and accurate method with a Single Shot MultiBox Detector will be used to detect the position of fruit, and a stereo camera to detect the two-dimensional position. The stereo camera will be attached to the gripper tool in order to obtain relative distance,orientation and size of the fruit. The harvesting stage requires the initial fruit location,the computation of the inverse kinematics of the robotic arm in order to place the gripper tool in front of the fruit, and a final pickup approach.After calculating the angles of the joints at the detected position by inverse kinematics, the robot arm will move to the target fruit’s position. The robot will then harvest the fruit by twisting the hand axis.

Project Objectives

The major objectives of our project are:

•The development of automatic systems that replace manpower in tasks while a person performs worse than an automatic device in terms of precision, repetitivity and working cycle

•To reduce the labor as well as cost

•To save the time and increase the accuracy

•To do effiecient work

And talking about our country which is a agro-based country and our whole economy depends on our agriculture. But if we look at our agricultural technology which is not still much advanced and the fruits are still picked manually.  Manual picking of fruits is not so effecient and requires more time and labor so why not replacing the labor with technology and get the work done effeciently.

Project Implementation Method

1. Research Methodology:

• Fruit detection and localization on trees using computer vision.
• Using color camera and a Single Shot MultiBox Detector (SSD) we 2D & 3D position of the fruit.
• Robot arm motion to the position of the detected fruit and fruit harvesting by the end effector without damaging target fruit and its tree.
• Block Diagram:

• 

• Computer vision.
• Image auquisition using color sensors and Image recognition.
• We will conduct fruit harvesting using a robot with a stereo camera installed below the gripper of robotic arm such that the fruit tree is

•  looked up from directly below.We will use universal robot as the robotic arm. The location of the fruits will be performed by matching fruit centroids instead of matching the complete stereo-image as a way to reduce matching complexity algorithm and foster the future development of real-time industrial systems.

• (SSD).

• Single Shot MultiBox Detector  (SSD) to detect the two dimensional  (2D) position of the fruit. SSD is one of the general object detection methods. SSD can comprehensively judge from color and shape. SSD speeds up the process by eliminating the need for the region proposal network. To recover the drop in accuracy, SSD applies a few improvements including multi-scale features and default boxes. SSD also detects 3D the position of fruits by measuring the  angles.

• Robotic Arm Motion:

• The control of the robotic arm requires an estimate of the fruit distances,positions and sizes in the trees in order to

  propose an automatic fruit harvesting procedure. A three-dimensional  (3D) position will be obtained to send a command to the robot arm. A stereo camera will be used to measure the 3D position of the fruit detected by the SSD.

• Inverse kinematics model:
• Inverse kinematics refers to the reverse process. Given a desired location for the tip of the robotic arm, what should the angles of the joints be so as to locate the tip of the arm at the desired location. There is usually more than one solution and can at times be a difficult problem to solve

Benefits of the Project

There are very tangible benefits of robotics on modern agriculture for producers and consumers alike. Some of the most significant benefits are:

Decreased cost of production;
Information sharing between producers and consumers;
Improved supply chain;
Increased labor efficiency and decreased cost;
Reduced food wastage and increased yields;
Increased financial sustainability.

Technical Details of Final Deliverable

Our final prototype of fruit harvesting robot will successfully pick the fruits with full accuracy of more than 90% and with a time less than 15 seconds.

Final Deliverable will be a robotic arm,which will be controlled by giving output through LCD.

The user will give input to the system that he wants to pick the red apples only, and then the system will check its database and by using machine learning,it'll gve instructions to the robotic arm to pick the red apples from the tree in real time. This project is a industrial level project and it has great innovation of picking multiple fruits in real time. We'll use python coding.

Final Deliverable of the Project Hardware SystemCore Industry AgricultureOther IndustriesCore Technology Artificial Intelligence(AI)Other Technologies RoboticsSustainable Development Goals Industry, Innovation and InfrastructureRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	55470
Raspbeery pi 4	Equipment	1	25500	25500
Raspberry pi camera	Equipment	1	9000	9000
ESP32	Equipment	1	850	850
L298n motor driver	Equipment	1	320	320
Ultrasonic sensor	Equipment	2	200	400
Car kit	Equipment	1	440	440
775 DC motors	Equipment	4	550	2200
Servo motors	Equipment	5	750	3750
Battery 3.7V 2200mah	Equipment	8	190	1520
Battery charging module	Equipment	2	60	120
Battery holder	Equipment	2	140	280
Buzzer	Equipment	1	150	150
Buck converter	Miscellaneous	2	740	1480
Robotic Arm kit	Equipment	1	1550	1550
OLED 1.9inch display	Equipment	1	500	500
MOSFET	Miscellaneous	4	50	200
NPN transistor	Miscellaneous	4	30	120
LED Mix colors	Equipment	20	230	4600
Resistors	Miscellaneous	30	20	600
Capacitor	Miscellaneous	10	20	200
Switch	Miscellaneous	1	150	150
Charger	Equipment	1	250	250
LDR	Miscellaneous	1	100	100
Charging connectors	Equipment	2	80	160
Wires	Miscellaneous	40	10	400
3.3V regulator	Miscellaneous	2	25	50
IR LED Transmitter	Miscellaneous	4	20	80
Photo Diode	Miscellaneous	2	250	500