Machine Learning and IOT based Plant Diseases Detection and Monitoring Robot
Agriculture plays a vital role in the development of the country. In Pakistan about 47% of the population depends upon farming. The global population is expected to reach 9.6 billion by 2050. In Pakistan the population has drastically increase day by day. One of the most challenging question
2025-06-28 16:34:03 - Adil Khan
Machine Learning and IOT based Plant Diseases Detection and Monitoring Robot
Project Area of Specialization RoboticsProject SummaryAgriculture plays a vital role in the development of the country. In Pakistan about 47% of the population depends upon farming. The global population is expected to reach 9.6 billion by 2050. In Pakistan the population has drastically increase day by day. One of the most challenging question the food and the agriculture industry is think how to double the food production with less cultivated land available every year. Against the challenges such as extreme weather conditions, raising climate change and environmental impact resulting from intensive farming practices. The demand for more food has to be meet. Diseases in plant are quite natural. If proper care is not taken in this area then it causes serious effects on plants and due to which respective quality and production is affected. The naked eye observation of the experts for detection of plant diseases is very costly and time consuming for monitoring in big farms which result in decreasing the quality of food. In order to eliminate the constant manual intervention, a smart farming monitoring system can built with the help of IOTs. The purpose of IOT in farming to monitoring a field with the help of agriculture sensors to predict the environmental parameters to analyze that how much food we have to produce for the next three months. Automatic detection and classification of plant diseases through machine learning algorithm is very beneficial for the farmers to detect the diseases at earlier stage so it increases the growth and productivity.
Project Objectives- To reduce the Human interaction in agriculture farming because it is less efficient and time consuming.
- To Facilitate the farmers by providing them pesticides free environment.
- To continuously monitoring the field data is provide a good statictics analysis of food production in agriculture domain.
- To improve the agricultural sector for better food productivity and quality.
- First we have made the mechanical body of the robot by using PVC pipes. We are using two window power motors for left and right sided. Our robot is rely on tri chain mehanical system in which two tyres are mechanical fitted with one power motor.
Secondly we have implemented the RF technology for the remote controlled robot movement. Path navigation can be done through IP cameras for live video Streaming. We have completed the transmitter section of robot for wireless data transmission. In transmitter Section, we collect analog data from 2D joystick module to arduino nano then nano has to be send data to Transceiver module NRF24L01. This wireless transceiver module have transmission frequency is 2.4GHz.
- In third phase we are working on Receiver section of the robot that is in progress. In Receiver part, we are using same Transceiver module for wireless data receiver and send to Arduino Mega. A monster Motor Sheild is attached to mega and two window power motors attached with full H-Bridge monster driver sheild.Our technical solution of RF communication section is given as:
- In Forth phase we are working on smart farming monitoring system based on IOTs technology. The purpose of IOTs in smart farming is to collect the environmental parameters data through sensors and send to cloud through MQTT protocol. Then we remotely access the data from cloud to Thingspeak IOT MATLAB Platform. Our technical solution is given as:
- In Fifth phase we are working on plant diseases detection through machine learning techniques. This part can be done in Matlab and our technical solution is given as:
Machine Learning Simulation Results are as :-
- It is very helpful in increasing the quality and quantity of the plants because it require less input.
- Farmer will have less interaction with the harmful pesticides which is better for their health.
- It is much attractive in digital farming because it provide them healthy environment.
- Its cheaper rate will be a reason for its popularity among farmers.
- To monitoring the status of the plants in the large field and collecting the data is very beneficial for the agriculture industry.
we have total three phases of our final technical soloution.
1. Wireless Communication Phase.
Through RF technology, we have acheived the low cost transmission with the larger freqiuency band 2.4 GHz. Through transceiver module we have received 6 transmission data at the same time. We have created a pipeline communication through transmitter and receiver in oder to secure our communication. So data is successfully transmitted and received in open environment. We are controlling our robot bu using IP cameras for path navigation.
2. IoT Phase
Through IoT technology, we easily collect the agricultural sensors data through nodeMCU ESP8266 wifi module. This module has to be created the packets of data and send over the cloud through MQTT and TCP protocols. We have to monitor the real time sensors data from cloud into the MATLAB ThingSpeak platform for the better visulaization. Also we have to take action on that data based on environmental parameters.
3. Machine Learning Phase
We are working on real time diseases detection and classification based on un-supervised learning. First we have to input the image through IP camera and then we have to apply some preprocessing techniques on that input data. Segmented part has to be done through K-Mean cluster algorithm. For classification and detection, we are using color based feature extraction such as HSI, RGB etc. After detection of the disease we generated a signal that will wireless transmit through bluetooth module that indicate the receiver section of the robot that turn on the pesticides spray.
Final Deliverable :-
We make a robot that move in the plant field to check the status of the plant by images shot by IP camera and send to MATLAB to identify the diseases based on real time. After the Diseases has been detected the robot is able to take action such as pesticides spray on plants. At the same time the real time monitoring has been done through agriculture sensors data to indicate the statictics analysis of food productivity based on environmental parameters. If any climate changes occurs the robot turn on the alarm system to indicate the farmers that they have to take some furthur steps to protect the plants from environmental changes.
Final Deliverable of the Project HW/SW integrated systemCore Industry AgricultureOther Industries Agriculture , Food , Energy , Health , Telecommunication Core Technology RoboticsOther Technologies Artificial Intelligence(AI), Internet of Things (IoT)Sustainable Development Goals Good Health and Well-Being for People, Affordable and Clean Energy, Decent Work and Economic Growth, Industry, Innovation and Infrastructure, Responsible Consumption and ProductionRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 79540 | |||
| Robot Body | Equipment | 1 | 30000 | 30000 |
| Plan PVC Socket | Equipment | 6 | 25 | 150 |
| PVC Pipe | Equipment | 1 | 950 | 950 |
| PVC pipe T-socket | Equipment | 18 | 50 | 900 |
| PVC pipe elbow-socket | Equipment | 26 | 60 | 1560 |
| PVC pipe Solution | Equipment | 1 | 550 | 550 |
| Metalic clip large | Equipment | 2 | 45 | 90 |
| 10 foot PVC pipe | Equipment | 10 | 40 | 400 |
| PVC small sockets | Equipment | 3 | 65 | 195 |
| Robot Tyres | Equipment | 4 | 325 | 1300 |
| 12V Battery | Equipment | 1 | 1650 | 1650 |
| 12V battery charger | Equipment | 1 | 1150 | 1150 |
| FeCl3 | Equipment | 1 | 180 | 180 |
| Metal Geras | Equipment | 6 | 300 | 1800 |
| Timing Chain | Equipment | 13 | 100 | 1300 |
| Cameras Stand | Equipment | 2 | 375 | 750 |
| Pavel | Equipment | 1 | 350 | 350 |
| Arduino Nano | Equipment | 1 | 400 | 400 |
| Transceiver module | Equipment | 2 | 700 | 1400 |
| Bluetooth module | Equipment | 1 | 500 | 500 |
| Joystick module | Equipment | 1 | 180 | 180 |
| Robot Body | Equipment | 0 | 0 | 0 |
| Plan PVC Socket | Equipment | 6 | 25 | 150 |
| PVC Pipe | Equipment | 1 | 950 | 950 |
| PVC pipe T-socket | Equipment | 18 | 50 | 900 |
| PVC pipe elbow-socket | Equipment | 26 | 60 | 1560 |
| PVC pipe Solution | Equipment | 1 | 550 | 550 |
| Metalic clip large | Equipment | 2 | 45 | 90 |
| 10 foot PVC pipe | Equipment | 10 | 40 | 400 |
| PVC small sockets | Equipment | 3 | 65 | 195 |
| Robot Tyres | Equipment | 4 | 325 | 1300 |
| 12V Battery | Equipment | 1 | 1650 | 1650 |
| 12V battery charger | Equipment | 1 | 1150 | 1150 |
| FeCl3 | Equipment | 1 | 180 | 180 |
| Metal Geras | Equipment | 6 | 300 | 1800 |
| Timing Chain | Equipment | 13 | 100 | 1300 |
| Cameras Stand | Equipment | 2 | 375 | 750 |
| Pavel | Equipment | 1 | 350 | 350 |
| Arduino Nano | Equipment | 1 | 400 | 400 |
| Transceiver module | Equipment | 2 | 700 | 1400 |
| Bluetooth module | Equipment | 1 | 500 | 500 |
| Joystick module | Equipment | 1 | 180 | 180 |
| Arduino Mega | Equipment | 1 | 1300 | 1300 |
| Water Pump | Equipment | 1 | 250 | 250 |
| Window power motors | Equipment | 2 | 1000 | 2000 |
| Monster Motor Driver | Equipment | 1 | 1800 | 1800 |
| Relay module | Equipment | 1 | 120 | 120 |
| Rain Sensor | Equipment | 1 | 250 | 250 |
| Air Pressure Sensor | Equipment | 1 | 200 | 200 |
| DHT22 sensor | Equipment | 1 | 450 | 450 |
| LDR module | Equipment | 1 | 120 | 120 |
| Buzzer module | Equipment | 1 | 100 | 100 |
| NodeMcu ESP8266 | Equipment | 1 | 700 | 700 |
| PCB copper | Equipment | 3 | 250 | 750 |
| Thesis color printing pages | Miscellaneous | 100 | 20 | 2000 |
| Thesis Binding | Miscellaneous | 5 | 850 | 4250 |
| Silicon Gun Glue | Miscellaneous | 1 | 1500 | 1500 |
| Silicon strips | Miscellaneous | 12 | 120 | 1440 |
| Double Tape | Miscellaneous | 1 | 800 | 800 |