Capsicum Harvesting Robot in Green Houses

Project Title

Capsicum Harvesting Robot in Green Houses

Project Area of Specialization RoboticsProject Summary

In modern agriculture, there is a high demand to move from tedious manual harvesting to a continuously automated operation especially in an Agriculture based country like Pakistan. The development of Capsicum Harvester with 92% success rate have already been performed at Queensland University of Technology (QUT), Brisbane, Australia. An approach to mechanize the agriculture sector of Pakistan keeping in view the scarcity of skilled labour, volatility in production due to uncertain weather events and high labour costs, introduction of robotic harvesting is a necessity. As Agricultural Engineers, our objective is to design a cost effective autonomous harvester in general and capsicum harvester in particular, customized to our environmental, socio-economic conditions and accessibility to farmers.

The project is aimed at designing, testing and calibration of visual servoing task of preferrably 3-6-DOF or 7DOF manipulator (6DOF articulated arm + lift joint), working model of Robotic Capsicum Harvester. The design consists of building an arduino robot from 3D modelled parts (if somehow it could not be managed to produce such 3D printed parts due to unavailability of 3D printing services in Pakistan then, these parts will be produced by CNC Lathe or Laser cutting). 

The project includes complete building of an arduino robot using arduino code, Servo motors, Stepper motors & drivers, Printed Circuit Boards (PCBs), Suction Cup, Microswitch, Magnets connecting suction cup to oscillatong blade, Robotic Wheels Platform, Vacuum pressure sensor & hose and RGB-D Camera (as its use may exceede the overall cost of the project so it can be replaced by cost effective sensors), ROS, MATLAB or C++ can be used to communicate between the hardware and software components.

As the robot is started in the field, the initial stage of the perception process involves scanning the crop row with the end-effector mounted RGB-D sensor. It scans the scene into single 3D model, using colour information and grasping poses through motion planning system.In another stage of perception, the sweet pepper from the background (leaves etc). Grasp poses for each capsicum are calculated using the segmented 3D point cloud of a capsicum. The grasp pose will place suction cup squarely on a planar region in the centre of the front face of the sweet pepper, while the cutting pose is calculated to be offset from the top face of the capsicum Harvesting trajectories are calculated relative to the grasping and cutting poses, attachment trajectory starts at a fixed offset back from the grasping pose and moves the suction cup along the selected approach axis. The end effector moves vertically from the attachment pose in order to decouple the suction cup from the cutting cup. The oscillating cutter cuts the crop, the detached capsicum falls from the plant and hangs freely from the flexible tether released into crate. The suction & cutter are magnetically recopled.

Project Objectives

The project consists of following objectives:

1. To develop the working model of robot for Capsicum Harvesting.
2. To enable the robot to grasp Capsicum of any shape and size so that this model can be applied to wide range of fruits and vegetables.
3. To ensure reduced labour peaks, labour costs and narrow down profit margin pressures.
4. To enable regular and selective harvesting, optimising crop quality, scheduling and therefore profit.
5. To introduce a cost effective harvesting technology which provides economics friendly solution, tailored to the needs of farming community.
6. If robotic-crop harvesting is to become a reality, there are three key challenges that must be addressed:

• Detection: determining the location of each crop
• Grasp Selection: determining the 3D pose and shape of each crop and selecting appropriate grasping and/or cutting points.
• Manipulation: detaching the crop from the plant without harming the crop or plant

Project Implementation Method

1. The basic parts of the robotic harvester arm may be obtained from 3D printing of the proposed shapes (if 3D printing is not viable then CNC Lathe work or Laser cutting will be performed).
2. The algorithm for robot harvester will be designed by using some effective Programming softwares such as ROS (Robot Operating System), MATLAB and/or C++.
3. The robotic arm is developed by assembling the parts, components and systems comprising of hardware and software in an integrated fashion.
4. The end effector body is adjusted as to efficiently perform the major harvesting operation by sensing and/or scanning through RGB-D sensor camera
5. The calibration of suction cup is done, which creates vacuum to hold the crop and the oscillating cutter cuts the crop.
6. There are following five major tasks to be performed by the harvester:

• Scanning: The robot arm is moved in pre-determined scanning motion to build up a 3D model using an eye-in-hand RGB-D camera
• Crop Detection: Sweet peppers from the 3D scene are segmented using colour information and localized by fitting a 3D parametric model to the segmented points.
• Grasp Selection: Candidate grasp poses are computed using the segmented sweet pepper point cloud.
• Crop Attachment: Suction cup grips the sweet pepper
• Crop Detachment: Oscillating blade cuts the sweet pepper from the plant.

Benefits of the Project

Following benefits are provided by robotic harvester to the farming community:

1. It will turn worsening farm labour shortage into automated timely harvesting.
2. With the increase in advancements in processor speeds, the robotics have paved the way to become more practical and cost effective.
3. It will lead to save the time in harvesting, high labour costs will be avoided.
4. It will ensure more precise and accurate harvesting of delicate fruits and vegetables.
5. These robots can work in extreme weather conditions as well where manual picking is not feasible.
6. It will revolutionize the horticulture industry by reducing the picking cost, maximizing the quality of produce and ultimately improving the sustainability of farming enterprises.
7. In an agriculture based country like Pakistan, mechanization is the need of the hour which directly benefits the agriculture sector.

Technical Details of Final Deliverable

The final deliverable form of the project will include the following:

1. The working model of the capsicum harvesting robot
2. A robotic arm with ranging between 3-6DOF
3. 3D modelled parts of the robotic harvester or else laser cut parts will be used in the development
4. Arduino based robotic arm, PCBs with installed servos and stepper motors, battery, gear box, computer hardware and robot controller
5. End effector body containing RGB-D camera, Vacuum Suction cup with hose and Oscillating cutter with magnets
6. Programming on the Robots Operating System, MATLAB, and/or C++
7. Robot mounted on wheels with complete robotic arm and end effector assembly

Final Deliverable of the Project HW/SW integrated systemCore Industry AgricultureOther Industries IT , Food , Energy , Manufacturing Core Technology RoboticsOther Technologies Artificial Intelligence(AI), Internet of Things (IoT)Sustainable Development Goals Decent Work and Economic Growth, Industry, Innovation and InfrastructureRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	41520
PCBs	Equipment	3	300	900
3D Robotic Arm Parts (metallic)	Equipment	1	13000	13000
Vacuum Suction Cup & vacuum hose	Equipment	1	1500	1500
Depth sensor camera	Equipment	1	10000	10000
Stepper Motor Nema 17	Equipment	4	1700	6800
Oscillating Cutter	Equipment	1	1500	1500
Servo Motors (SG90)	Equipment	4	1200	4800
Servo Motor (A4988)	Equipment	1	220	220
Servo Motor (MG996R)	Equipment	1	900	900
Vacuum sensor	Equipment	1	1900	1900