Agricultural Drone
Deforestation is one of the major issue behind all the global warming and increasing heat waves because scientifically the Plants do photosynthesis in which they consume CO2 (Carbon Dioxide) which is produced by chemical industries, Petroleum based Vehicles, Atomic Reactors, Human and Animals
2025-06-28 16:25:02 - Adil Khan
Agricultural Drone
Project Area of Specialization Artificial IntelligenceProject SummaryDeforestation is one of the major issue behind all the global warming and increasing heat waves because scientifically the Plants do photosynthesis in which they consume CO2 (Carbon Dioxide) which is produced by chemical industries, Petroleum based
Vehicles, Atomic Reactors, Human and Animals (as a result of respiration exhales CO2) etc. The CO2 gas is warm in nature on the other hand the Oxygen gas is cool in nature so If plants stops consuming CO2 gas the temperature will get higher in that environment so cutting of plants and less irrigation of plants causing higher rate of CO2 in our climate gradually increasing temperature and less fresh air (less concentration of oxygen gas) and in urban areas the rate of pollution is very high
Now a day due to destruction of plants and deforestation in urban areas, the rate of pollution is increasing day by day so we need this type of technology to grow large number of plants as soon as possible to control the rate of pollution. Forest regeneration is fundamental to sustainable forestry; and often, direct seeding is a lowcost and dependable method of creating desirable forests. Direct seeding can be performed manually, with ground-based machinery or from the air
In this project, we decided to introduce a drone which will be climate compromised (No CO2 production while using machine) a combination of IOT & AI based product that contains web application as a controller and soil detection model using image recognition in the area which is affected by global warming, less agricultural or a deforested land. The purpose of this project is to develop a “Smart Drone” that is capable of dispersing seeds in above affected fields with potential for reforestation which helps in spreading seeds over those areas where human being is unable to plant trees or human is not only enough to do planting due to humanize nature (I mean human consumes more time and more energy to do this). So, the basic working of drone will be as follows:
The drone will be first connected with a “custom web application” (Drone Controlling App) and shows it hardware status to the User.
When the status is green on the web interface then user can see the camera view of the drone in his device and when the user gives commands the Drone will then fly up in the sky and will try to gain stability at specific height once it becomes stable in air it will start “Image detection process” using camera in which it will find for the soil on the surface area of earth and will try to identify which part of earth is soil or not once the condition becomes true the drone will drop the special seeds which can be grow with less water.
we used to develop the drone and the drone will be composed of two parts first is controller and second is soil detection through image processing and temperature detection.
Custom Drone:
The proposed drone assisted seed dropping system will provide an essential tool for an efficient reforestation. Due to the effectiveness, this system is also suitable for reforestation in ecosystem where germination has not occurred.
The purpose of the project is to develop a motor controlled autonomous flying vehicle that is capable of dispersing seeds in agriculture fields with potential for reforestation. This work includes two components: a custom-build Quad copter and software (Web based) that can generate GPS coordinates for mission trajectory, control the seed dispenser, and communicate with the operator(s). with RTL return to launch mode to complete mission and again go to home location.
Custom Software with A.I. Features:
Our customized drone used with web application for controlling. Once a drone fly then it senses or get temperature according to geolocation and if the temperature is high in areas then it takes image of land then send it to the A.I. model_v1 and the model_v1 will then tell whether the taken image is a land or not basically soil detector once the land is detected then the image will be send to the next A.I model_v2 and the model_v2 will classify in to fertilize soil category tested by kaggle dataset and the system will tell user to which class the soil belongs to whether the soil is fertile or not and our drone is used in those areas where the rate of pollution is high and the land is affected by global warming so we used this drone for spreading seeds to reforest our country. It also useful for farmers for planting the seeds it saves the cost and time because human consumes more time and energy to do this work.
IOT:
The Internet of things that are used in physical objects that are embedded with sensors, processing ability, software, and other technologies, and that connect and exchange data with other devices and systems over the Internet or other communications networks.
In our project we used IOT for embedded the software in hardware device. We used IOT technology in this project for integration of software in hardware device and link the system we used some sensors for detection so used this technology.
Computer Vision:
In computer vision it involves the study and development of algorithm and achieve automatic visual understanding. We used some concepts of machine learning and used keras permade algorithm in Tensor Flow library.
Project Implementation MethodWe are going to adopt AGILE Methodology to completion of the project before discussing the reason to choose agile methodology. First we need to know what Agile Project Management is.
There are mainly six stages in Agile Development life cycle:
1. Project Requirements
The top priority step in agile methodology is Requirements gathering phase ,In order to gather the requirements of our Agricultural Drone project we are intended to arrange meet ups with many senior, drone vendors, environmental protection departments, seed balls manufacturers, hardware manufacturers to gather hardware related requirements, software developers
2. Project Design
In Agile methodology these all Functional and non-functional requirement defines by UML, DFD’s or Class Model Diagram.Further iterations are spent refining the initial design and/or reworking it to suit the new feature actually this is the beauty of agile methodology.Temperature sensing, soil fertility detection, GPS Location tracking, Seed Firing, Seed Balls Manufacturing, Use of RasberryPI, Image Processing, Modules integration etc. So overall Agile Methodology is fit for our project
3. Analysis
In this phase we already have all functional and non-functional requirements, Software and hardware related requirements After everything setup and in place hardware and software compatibility will be considered if any inconsistency occurred during the analyzing process iteration cannot be proceed further until all requirements of the sprint satisfied successfully.
4. Integration and Testing
This one is the most difficult and crucial phase, in which components (software and hardware) are combined to confirm that they interact according to our expectations and requirements. This phase makes sure that software is bug-free and compatible with hardware and everything else. During further iterations of this SDLC stage, the more testing involves and not only consider the functional testing but also the system integration testing. In order to perform integration and testing you need an environment that closely mirror the target without an environment integration test result will be ambiguous. The test environment should mirror the hardware and software configuration that the project will run when delivered as closely as possible to ensure that the results you see are predictive.
5. Implementation
Application/Website is deployed on the server and run with the Drone with real time environment.
6. Review
Once all the development phases successfully complete team once again review the progress whether it fulfill the requirements or not. Also the Team makes consultations towards resolving the problems that arose during the previous phases. Afterwards the Agile Software Development Lifecycle phase starts a new ITERATION.
Project Model Diagram:
System Design and Architecture:
The benefits of this drone are vast. This is going to be great in restoring our ecosystem. For example, if a farmer spread 100 seeds on daily basis. This Smart Drone can fire up to 1,000 seeds on daily basis that means in a year we can fire up to 365000 seeds that means we can grow and build more and more artificial forests. This drone can do the work of 10 labors at a time. The cost of farming will decrease gradually.
When the project will complete so it will be used in agriculture fields. Farm Engineers use this drone for large number of farming the seeds in fast and efficient manner and it also use for growing plants in that type of places where the rate of pollution is high. Aerial seeding in the USA has been successfully applied on areas after storms or fires where ground vehicles were unsuitable due to stumps and other obstacles. Almost 75 % of the aerial seeding was done with the use of manned aircraft and helicopters. In areas for reforestation in excess of around 200 ha, aerial seeding comparable to the cost of most of the ground broadcast methods of sowing, and allows one to complete the work in a short time
As an Engineers we can use all our innovation and all our ideas to develop this world in a better place. The challenge that we’re tackling is a complex one and working with a passionate team. Came up with this idea to use automation and digital intelligence to plant billions of trees. The project aims to develop a drone that is capable of dispersing seeds in agricultural fields.
The scale of land degradation today has gone too far. We can no longer just rely on traditional methods to restore the land we have to increase the rate. And we have a target of planting 500 billion trees. There’s a saying that goes that the best time to plant a tree was 20 years ago and the second best time is Now. We have this opportunity now and we need to act today.
By using Seed dropping drones, the scale of what we’re going to do is one which is going to re-green. Restoration of forests within landscapes offers multiple social, economic, and environmental benefits that enhance lives of local people, mitigate effects of climate change, increase food security, and safeguard soil and water resources.
Climate change makes every other problem harder to tackle. We are going to resolve to find a way to mitigate it.?
The proposed drone assisted seed dropping system will provide an essential tool for an efficient reforestation. Due to the effectiveness, this system is also suitable for reforestation in ecosystem where germination has not occurred.
Technical Details of Final DeliverableFunctional Requirement:
Getting Temperature:
The drone controller is getting the data from web API so it can measure the temperature to identify weather the needs of planting in this particular area or not to overcome increasing the warming day by day due to lack of trees.
Soil Detection:
Arial images will obtain by flying a small drone over the area of interest (by using Image recognition with Machine Learning on Python, Image processing we identify the fertility of soil) to analyze the soil conditions, terrain conditions, moisture content, nutrients content and fertility levels of the soil. Accurate Detection of the Soil Steps:
• Conversion of Color Model
• Image Segmentation Based on Color
• Converting the Segmented Image into Grayscale Image
• Splitting of Gray Scale Image to Consider Spatial Heterogeneity of Light Intensity
The captured photographs are models of RGB color space. RGB color space presents the large number of colors by mixing various proportions of red, green and blue. However, HSV (hue, saturation and value) color space is more appropriate for the detection of various shades of the color as compared to RGB. Hence, conversion of image from RGB to HSV color model can be an effective solution. This is done using OpenCV built-in function.
A novel Python script was demonstrated in this study to automate the procedure of soil color analysis for interpreting surface moisture content. Unlike the conventional approach, the novel technique considers entire surface area of soil and spatial no uniformity of light intensity to interpret soil moisture content. The results obtained by the novel approach are reproducible and computation time is low. The difference between the brightness values (mean gray values) obtained by conventional manual image processing method and newly developed technique is less than 3%. The newly developed Python script can be programmed into unmanned air vehicles (UAV). These UAVs can be directly connected to the irrigation systems to control watering. Hence, unmanned irrigation maintenance system can be established for geotechnical and green infrastructure systems using the newly developed Python script.
Non-Functional Requirements:
• Maintainability
• Efficiency
• Reliability
Web Applications Section for Users:
• The web should be responsive.
• The web should be secured and authorized for users.
| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 79919 | |||
| Raspberry Pi 3 b+ | Equipment | 1 | 28000 | 28000 |
| Brushless Motor | Equipment | 4 | 2000 | 8000 |
| Raspberry pi Camera | Equipment | 1 | 800 | 800 |
| F450 Quad rotor Frame with integrated PCB. | Equipment | 1 | 14000 | 14000 |
| Esc Controller | Equipment | 4 | 800 | 3200 |
| APM Module 2.8 flight controller | Equipment | 1 | 7119 | 7119 |
| Propeller | Equipment | 4 | 100 | 400 |
| Dry Cell Battery | Equipment | 1 | 6000 | 6000 |
| Node mcu | Equipment | 1 | 1000 | 1000 |
| Mpu6050 | Equipment | 1 | 1000 | 1000 |
| Charger | Equipment | 1 | 400 | 400 |
| Casing, Printing of standee, flyer, poster, report,Travelling | Miscellaneous | 1 | 10000 | 10000 |