Chettinad Vidyashram Science Expo 2025

Crop Yield Prediction using Machine Learning and Robotics

Revolutionizing agriculture with machine learning and robotics, delivering data-driven insights and automation for sustainable farming

Project Overview

How It All Works

The system operates on a simple but powerful flow, connecting hardware, software, and intelligence to optimize farming practices.

Step 1: Sensors - Soil Moisture, DHT11 (Temp & Humidity), and LDR (Light) collect real-time data.
Step 2: Arduino Uno - The microcontroller collects sensor readings and sends them to the computer.
Step 3: Machine Learning Model - A Decision Tree model analyzes the data to predict yield conditions and make intelligent decisions based on soil moisture ranges.
Step 4: Dashboard - A Python GUI visualizes sensor values, provides a control interface for the farmer, and gives voice alerts.

AI vs. Manual Monitoring

Traditional farming relies on a farmer's observation and guesswork, which often leads to resource waste. Our AI-powered system uses real-time data and learning algorithms to make smarter, more precise decisions.

AI uses in-field sensors and ML to analyze data continuously.
It learns patterns to make smart irrigation and input decisions.
Reduces water waste and increases yield accuracy, which manual methods can't match.

Basic Model Explanation

Our Decision Tree model uses soil moisture levels to classify plant health and predict yield conditions, ensuring your crops get exactly what they need.

Soil Moisture < 20% or> 90%: Unsustainable
Soil Moisture 20-40%: Low Yield
Soil Moisture 40-60%: Medium Yield
Soil Moisture 60-75%: High Yield

Robotics - The Hands & Legs

Our robot system acts as the "hands and legs" of the farm. It doesn't spray blindly but checks soil moisture from sensors. If the soil is dry, the robot sprays; if it's wet, it skips, saving water.

Water-saving automation.
With smart nozzles and section control, it only waters areas that need it, preventing waste.
No overlap, no waste.

Machine Learning - The Brain

Machine learning functions as the "brain" of the system. Instead of fixed rules, it studies patterns from sensor data, such as when soil usually dries, and how weather affects water loss.

The ML model "learns" from data over time and improves its decisions.
For example, it might realize "Plot A dries faster than Plot B, so water Plot A more often".
This leads to smarter irrigation schedules and higher yields.

API + Python Interface - The Face & Voice

The API is a messenger that ensures sensors, the robot, and the ML brain talk to each other smoothly. The Python UI is the farmer's dashboard , providing a visual and audio alert system.

Shows real-time data like soil moisture, temperature, and plant stress.
Voice alerts, such as "Need Water," notify the farmer when crops are at risk.
Displays maps and charts of the field, and allows the farmer to control or override settings.

In-field IoT Applications

Smart sensors placed directly in the field continuously measure crop, soil, and micro-climate conditions and send the data to a central system via Python code.

Soil Moisture - tells how much water is in the soil, helping with irrigation scheduling.
pH Sensors - measures soil acidity/alkalinity for crop growth.
Micro-weather sensors (DHT11) - measure temperature, humidity, wind, and sunlight to help the ML model predict water needs.

Key Features

Machine Learning - The Brain

Learns from soil and weather data, creating smarter irrigation schedules for higher yields.

IoT Sensors - The Eyes & Ears

In-field devices track soil moisture, pH, and micro-weather in real time for precise decisions.

API + Python Dashboard

Real-time data, voice alerts, and interactive maps help farmers monitor and control their fields.

Decision Tree Model

Classifies soil moisture levels into yield predictions, ensuring accurate water management.

AI vs Manual Monitoring

Replaces guesswork with data-driven insights, reducing water waste and boosting yield accuracy.

Robotics - The Hands & Legs

Smart robots spray only when and where needed, preventing overlap and saving precious water.

Our Team

Meet the brilliant minds behind this innovative project

Meghna

Sagarika

Sravanthi

Anichamalar

Leksha

Reshma

Hasini

Dhisshan

Ananthashri

Harini

Sanjana

Deshna

Rithenya

Shamitha

Keerthana

Jahnavi

Rinith

Rahul

Karnesh

Tejas

Rithik Sai

Naveed

Aqeel

Nithya

Nadhira

Pranavi

Tiana

Aditi

Samyuktha

Darsh

Viswak

Harshik

Jerosh

Ruwayda

Dhananjeyan

Rohith

Lipika

Sahana

Rakshan

Harkishan

Prathick

Tatva

Harshik Kumar

Shraddha

Yasaswini

Sakshi

Sagithya

Kashvi

Ayanthika

Ayesha

Jeet

Project Status

Real-time progress tracking of our development phases

1	Front end	-	Functioning as expected	Modification in progress
2	Arduino Setup	-	Completed	Completed
3	Model	-	Completed	Testing in progress
4	Front end ML code	-	Completed	Tuning in progress
5	Integration	-	R&D in progress	In Progress
6	Testing	-	Existing functionality working fine	Working
7	Tuning	-	In progress	In Progress
8	Deployment	-	-	Pending