Problem Statement
The current agricultural scenario includes a lot of occupational injuries during harvesting and the lack of mechanization leading to lower harvest efficiency. For this project, we designed and fabricated a robotic harvester that would climb trees and harvest their fruit. The aim was to alleviate work-related risk to life while improving the throughput of the plantation.
Design Challenges
-
To make the product modular and flexible to varying tree girths
-
Design replaceable modules for easy maintenance
-
Design for structural integrity while keeping costs low
-
Design a lightweight and low power solution for harvesting fruit
-
Design an energy efficient power and control system
My role
Key Skills
Other Team Members
-
Project Lead
-
Designed and tested early prototypes
-
Designed the primary structure of the robot
-
Designed the control systems and on-board electronics
3D Design
Usability Testing
Hardware Prototyping
Microcontrollers
Smt. S Dawnee | Project Mentor | MSRIT, Bangalore
Abhinav Roy | MSRIT, Bangalore
Aditya Patil | MSRIT, Bangalore
Kickoff
In this project, we started out by identifying our target users and application. We chose harvest over pesticide spray as our application. The next step involved thorough research on existing mechanical harvest methods and the reasons for their failure. Selection of the final design took several iterations and the design thinking will be explained ahead. We started out by asking ourselves some initial key questions.
"What is the product and who is it for?"
"What is the most important need of our users?"
"How do we make the system modular and lightweight?"
"How do we implement low power harvesting"
"What challenges could we face moving forward?"
"How do we make the production scalable?"
We picked coconut and arecanut trees as our target trees as harvesting these trees posed the greatest risk to life. Also, these trees had near constant girths and smoother bark.
We decided on using an additive method of manufacturing over a subtractive method in order to reduce wastage and costs. The advantages of subtractive manufacturing, namely heat resistance and smooth surface finish, were not important to us.
3D printing using PolyLactic Acid (PLA) was our preferred choice due to its superior strength, lightweight and eco-friendly nature.
Project Flow
Competition Analysis
We looked at competitor's attempts at solving the problem, and although all of them achieved the objective of climbing a tree, most of them were not suitable for practical use. In addition, they were not capable of fruit harvest after climbing trees.
The major points of failure of these products were:
- Too bulky, difficult to transport and assemble
- Too slow, complicated mechanism
- Too many moving parts leading to frequent breakdowns
- Not adaptable to different trees
SOME OF THE COMPETING PRODUCTS
Our Solution
The first requirement of the robot was to overcome gravity so that the robot could remain attached to the surface of the tree. In order to climb the tree, a high torque was required, which could be provided by a gear train mechanism. The proposed design used an epicyclic gear system as shown in the figure below.
An epicyclic gear consists of three main gears: a sun gear at the centre, planet gears that rotate with respect to the sun gear, and a ring gear in which the sun and planet gears are contained. The advantage of an epicyclic gear is that it can provide very high torque for a very small size. Therefore, the robot could be built small, compact and lightweight.
The robot was designed to be modular and could therefore easily adapt to different tree girths. This also made maintenance quick and easy.
EPICYCLIC GEAR SYSTEM
CHALLENGE 1
Modular Product
Our goal was to make the harvest process efficient. That included reducing down time due to repair. Another important requirement was to make the robot adaptable to different tree girths. Therefore, we decided to divide the robot into multiple identical segments as shown in the image. This made it flexible to varying tree girths and makes repair quick and easy.
CHALLENGE 2
Structural Integrity at Low Costs
As the application would induce a lot of wear and tear on the parts, especially the gear teeth, we needed the robot to be strong. Most of our choices ensured structural integrity but would have made the robot heavy and expensive. Therefore, we chose 3D printing with PolyLactic Acid (PLA) as our material. It was strong, inexpensive and had high plasticity. An added benefit was that it was biodegradable.
CHALLENGE 3
Energy Efficiency
The robot was powered by a 2200mAh Lithium-Polymer (LiPo) battery pack which was lightweight and had high current density. This enabled the robot to climb and harvest approximately 30 trees per charge. The robot could be controlled wirelessly by a NodeMCU microcontroller and a joystick. This low power solution allowed a range of 300 metres.
CHALLENGE 4
Harvesting
A motor actuated cutter was considered for harvesting. But that made the robot heavy and required a lot of power. Therefore, we considered a nichrome hot-wire as shown in the image. It could be heated to 450º C at low power and could cut through branches with ease.
Key Takeaways
The robot was successfully able to climb and harvest the target trees.
-
It could harvest an average of 15 trees on a single charge.
-
The delay in response between the controller and the robot was only 200ms.
-
The farmer could cover the cost of investment in the product in around 45 days.
