Smart Environment Monitoring System

The Smart Environmental Monitoring system will not only simplify and streamline plantation management operations but also enhance efficiency and promote sustainability.

Smart Environment Monitoring System

The Smart Environmental Monitoring system will not only simplify and streamline plantation management operations but also enhance efficiency and promote sustainability.

Smart Environment Monitoring System

The Smart Environmental Monitoring system will not only simplify and streamline plantation management operations but also enhance efficiency and promote sustainability.

Smart Environment Monitoring System

The Smart Environmental Monitoring system will not only simplify and streamline plantation management operations but also enhance efficiency and promote sustainability.

Smart Environment Monitoring System

The Smart Environmental Monitoring system will not only simplify and streamline plantation management operations but also enhance efficiency and promote sustainability.

Introduction

Smart Environment Monitoring

Welcome to our innovative Smart Agricultural Monitoring System, a seamless blend of technology and sustainability. Our project focuses on remotely monitor critical environmental factors essential for crop health, such as:

Temperature
Humidity
Light Intensity
Soil Moisture
Location

The system collects real-time data from sensors placed in the field, transmits it to Firebase Firestore for storage and analysis. With this setup, managers can access live data through a user-friendly Mobile interface, enabling them to make data-driven decisions. Our solution aims to enhance agricultural productivity, reduce resource waste, and promote sustainable farming practices.

Project Repository

Solution Architecture

System

Our system is an IoT-based smart agriculture solution designed to monitor and analyze environmental conditions like temperature, humidity, light intensity, and soil moisture. It collects real-time data using sensors connected to an ESP32, transmitting information through LoRa and storing it in Firebase for analysis and visualization.

Highlevel System

The high-level system architecture includes interconnected components: sensors gathering data, ESP32 acting as the main controller, LoRa nodes for long-range communication, and Firebase as the backend for real-time data storage. The user interacts with the front-end web interface to view insights and predictions, while ML models process data trends for informed decision-making.

Implementation

Implementation involves integrating hardware and software components. Sensors are connected to the ESP32 for data collection, with LoRa nodes facilitating long-range data transmission. The ESP32 uploads data to Firebase via Wi-Fi, where it’s processed and displayed on a dynamic web interface. The ML model, trained using sensor data, predicts environmental trends, helping optimize plantation management.

Software Design

Mobile App

Mobile User Interface

The user interface is a web-based dashboard that visually displays real-time sensor data, including temperature, humidity, light intensity, and soil moisture levels. It offers an intuitive layout with graphs, maps, and alerts for easy monitoring. Mobile App specifically design for easy use while in the site and see data in real time.

Real-time data visualization
Interactive graphs for trends
Map integration for sensor locations
User-friendly and responsive design

Mobile App Technology Stack

The Mobile app frontend is built using Flutter for its component-based architecture and responsive design. The interface fetches data from Firebase and displays it in real-time with dynamic updates.

Responsive Design
Modern UI Components
Optimized for Speed

Web app

Web User Interface

The user interface is a web-based dashboard that visually displays real-time sensor data, including temperature, humidity, light intensity, and soil moisture levels. It offers an intuitive layout with graphs and maps for easy monitoring. Web app specifically design for managers and higher level positions to analyse data easily and clearly. Easy to handle administration role task effectively.

Real-time data visualization
Interactive graphs for trends
Map integration for sensor locations
User-friendly and responsive design

Web app Technology Stack

The Web frontend is built using React with Vite for fast development and optimized performance. Its component-driven architecture enables modular UI design, while real-time data is fetched and displayed dynamically from Firebase, ensuring a seamless user experience.

Responsive Design
Modern UI Components
Optimized for Speed

Backend

Database

Firebase Firestore is used as the cloud database to store real-time sensor data, including timestamps, coordinates, and environmental readings. It provides scalable, NoSQL data storage with automatic syncing.

Firebase Firestore (NoSQL cloud database)
Real-time data storage and syncing
Timestamp-based entries for sensor data
Scalable and secure

Backend Technology Stack

The backend uses Firebase Functions to process incoming data from ESP32 devices. It handles CRUD operations, ML model integration, and real-time updates to the frontend.

Firebase Functions for serverless processing
ESP32 for data collection and transmission
HTTP & MQTT protocols for data transfer

Hardware Design

Sensors and Modules

We chose following sensor according to the required accuracy level.

ESP32 Microcontroller
DHT22 Sensor
Hall Sensor
Neo-6M Sensor
Capacitive Soil Moisture Sensor
NPK Sensor
BH1750 Light Intensity Sensor
SD Card Module
RTC Module
DC to DC step down module
Rechargeable Battery

Main Circuit Diagram

Our Main Circuit Diagram illustrates the core components of our Smart Agricultural Monitoring System built around the ESP32 microcontroller. The system integrates multiple sensors, including the DHT22 for temperature and humidity, BH1750 for light intensity, and a soil moisture sensor to assess soil health. The Neo-6M GPS module provides real-time location data, while the hall sensor detects magnetic field changes, useful for monitoring mechanical operations. An LCD display is connected to visually present sensor readings directly. All components are powered by a battery source, with proper wiring ensuring stable communication through I2C (SDA, SCL pins) for the BH1750 and LCD, and GPIO pins for the remaining sensors. This setup enables seamless data collection and transmission via the ESP32, forming the foundation for real-time, remote agricultural monitoring.

Prototype

Prototype Implementation

Our prototype is a smart hardware-software integrated system designed to collect environmental or agricultural sensor data using microcontroller-based devices. The devices gather readings such as temperature, humidity, light, soil nutrients, and GPS coordinates, and log them locally to an SD card when offline, or directly upload to Firebase Firestore when WiFi is available. The system is built for reliability, featuring real-time data updates, error detection, and recovery mechanisms.

3D Design

Final Product

Portable Device Demonstration

Fixed Device Demonstration

Performance

Request Latency

Low sensor-to-cloud delay
Fast mobile/web response
Efficient ML model inference

Scalability

Dynamic load handling
Cloud-native architecture
Stateless components

Availability

Network Failure Handling
Redundant cloud deployment
Fallback mechanisms

Security

Secured API endpoints
Firebase Authentication
Encrypted data storage

Testing

We did

Software Testing

We did

Hardware Testing

We did

Machine Learning Training

We did

Backend Testing

Why Testing is necessary?

The main benefit of testing is the identification and subsequent removal of the errors. However, testing also helps developers and testers to compare actual and expected results in order to improve quality. If the software production happens without testing it, it could be useless or sometimes dangerous for customers.

Software Frontend Testing

Focuses on ensuring the user interface works seamlessly. This includes verifying correct data visualization, responsive design across devices, and smooth user interactions, ensuring real-time sensor data displays without glitches.

Hardware Testing

Involves checking the proper functioning of sensors (DHT22, BH1750, soil moisture, GPS), ensuring accurate data collection. Tests include sensor calibration, power stability, and connectivity validation with the ESP32 and LoRa modules.

Machine Learning Training

Trains a machine learning model using collected sensor data to predict trends like soil moisture levels or plant health. Testing ensures the model’s accuracy, checks for overfitting, and validates predictions using real-world data.

Backend Testing

Focuses on validating data flow between ESP32 and Firebase Firestore. Tests check for secure data storage, real-time updates, proper authentication, and verifying CRUD (Create, Read, Update, Delete) operations work as expected.

Timeline

Budget

Product Implementation

We implemented our product in Palm Oil Plantation and tested successfully with our stakeholder. The student team successfully developed and demonstrated prototypes for monitoring key agricultural parameters, including: