Title: HexaAgro: AI-Driven Hexacopter for Precision Pest

Management and Eco-Friendly Agriculture.

Student Investigators:
Student Members: 1. Riddhiraj Chanda 2. Ritam Das 3.Monodip Roy 4. Md Ayan Akhtar
Khan University of Engineering and Management, Kolkata

Faculty Investigators:
Dr. Subhajit Das, Senior Member, IEEE, Associate Professor, Department of ECE,
University of Engineering and Management, Kolkata.
Prof. (Dr.) Abir Chattopadhyay, Senior Member, IEEE DEAN (RESEARCH), HOD
(ECE), University of Engineering and Management, Kolkata

Keyword: Ultrasonic Pest Control, Mobile Agricultural Rover, Eco-Friendly Pest

Repellent, Function Generator, ESP32-CAM Surveillance, Smart Farming, Arduino-Based
Automation, Frequency Modulation for Pest Repulsion, Sustainable Agriculture.

Background and Expertise

Our team is composed of multidisciplinary individuals with expertise in:
1. Electronics and Embedded Systems – Designing and implementing sensor-based
systems.
2. Very Large Scale Integration (VLSI)—Designing and implementing the systems
using the concept of Surface- Mount Technology.
3. Data Analytics – Data collection, processing, and predictive analytics.
4. Mechanical Engineering – Prototyping and system integration for drones.
5. Computer Science – Development of algorithms for data processing and machine
learning.
6. Project Management – Ensuring timelines and deliverables are met.
Motivation
The project aims to revolutionize pest management in agriculture by integrating drone
technology with advanced AI-based pest identification and ultrasonic repellents. Current
methods often rely on chemical pesticides, which pose environmental and health risks. By
leveraging autonomous drones equipped with sensors and AI, this initiative provides a
sustainable and efficient solution, addressing critical challenges in eco-friendly bio-
agriculture. This project bridges the gap between traditional pest control methods and
innovative, eco-friendly technologies, making it both timely and impactful for modern
agricultural practices.

Project Summary:
The proposed project, HexaAgro, introduces a eco-friendly green-solution for precision
pest management in agriculture, leveraging the synergy of hexacopter drones, artificial
intelligence (AI), and ultrasonic repellent technology. Building on the success of the
AgroRover—a ground-based ultrasonic pest repellent system—this initiative transitions to
an aerial platform to address the challenges of pest control in tall crops, fruit orchards, and
extensive agricultural fields.

At the core of this innovation lies a hexacopter equipped with high-resolution cameras and
generative AI algorithms capable of real-time pest identification. This AI system is trained
to accurately classify pests and map them to specific ultrasonic frequencies, enabling
targeted pest repulsion. By replacing chemical pesticides with ultrasonic solutions, the
project promotes eco-friendly pest management while safeguarding human health and
environmental sustainability.

The drone platform offers unparalleled flexibility and scalability, with advanced navigation
capabilities for effective coverage of diverse terrains. Its modular design incorporates high-
precision sensors, a robust AI framework, and a customizable ultrasonic frequency
generator, ensuring adaptability to various agricultural contexts.

This project’s scientific foundation integrates cutting-edge technologies:

• Generative AI: Real-time pest classification and behaviour mapping.

• Ultrasonic Technology: Frequency-specific pest repellents that are species-

targeted and environmentally safe.

• Hexacopter Mobility: Efficient aerial deployment for comprehensive field

coverage and reduced tedious human-effort.

With a development roadmap spanning research, prototyping, testing, and field

deployment, HexaAgro is poised to revolutionize pest control practices. The expected
outcomes include significant reductions in crop damage, enhanced yields, and minimized
reliance on chemical pesticides. By combining innovation, sustainability, and scalability,
HexaAgro aligns with global goals for sustainable agriculture and promises a
transformative impact on modern farming practices.

Introduction:
Pests cause significant damage to agricultural produce globally, and chemical pesticides
remain the primary mitigation strategy, leading to environmental and health concerns. The
AgroRover addresses these issues with ultrasonic technology, but it is limited to ground
mobility, making it less effective for crops like sugarcane, fruit orchards, or large farmland.

The next step is transitioning to an aerial solution—a hexacopter equipped with a high-
resolution camera and Generative AI. This solution will identify pests in real time and emit
corresponding ultrasonic frequencies to repel them effectively. The system combines
sustainability, technological innovation, and adaptability, positioning it as a game-changer
for precision pest control.
Project Description:
Our already Developed Model: Ultrasonic Agricultural Pest Control Rover:

The "Ultrasonic Agricultural Pest Control Rover”

integrates mechanical, electrical, and electronic
components, ensuring a robust and functional design
suitable for agricultural applications. The system
design is divided into the following components:

Mechanical Components

1. Four-Wheeled Chassis:
- A sturdy 4-wheeled chassis forms the base of the
Fig1: Lateral View of Our Current Model
rover, providing stability and
mobility across uneven terrains found in agricultural
fields.
- The wheels are driven by high-torque DC motors
capable of handling the rough
environment.
2. Motors:
- Two DC motors are used for forward and backward
movement.
- The pan-tilt mechanism is driven by servo motors,
enabling the ESP32-CAM module
to rotate and tilt for optimal field coverage.
Fig 2: Shows the Rover Structure in
3. Pan-Tilt Mechanism: Side view consisting of all the devices
- Mounted on the chassis, this mechanism allows the on board.
ESP32-CAM to adjust its angle of
view for efficient monitoring and navigation.

Electrical Design
1. Arduino-Based Function Generator:
- The Arduino is programmed to generate
square waves in the frequency range of 100
Hz to 65 kHz.
- Frequency adjustment is facilitated using a
100k potentiometer, ensuring precise
control over the ultrasonic signals emitted.
2. ESP32-CAM Module:
- Serves as the rover's control and navigation
system, providing real-time video
Fig 3: Shows the signal Generator streaming and remote control via a mobile or
web interface.
- It also enables monitoring of the system’s operation and surroundings.
3. OLED Display (128x32):
- Displays the current ultrasonic frequency generated by the Arduino-based function
generator, allowing the user to adjust the frequency accurately based on pest requirements.

4. Power Supply:
- A rechargeable battery powers the entire system, ensuring extended field operation.

Project Description: Proposed Model

The proposed model introduces an advanced hexacopter-based system designed to
revolutionize pest management in agriculture. This drone system combines cutting-edge
AI-driven pest identification with targeted ultrasonic repellent mechanisms, offering a
scalable, efficient, and eco-friendly solution. Unlike ground-based systems limited by
terrain, the hexacopter provides superior mobility and adaptability, ensuring effective pest
control in diverse agricultural environments such as tall crops, orchards, and expansive
farmlands.

Core Components and Features:

1. Hexacopter Platform:
o Mobility and Payload Capacity: A robust hexacopter drone with high thrust
brushless motors, advanced electronic speed controllers (ESCs), and durable Li-
Po batteries ensures seamless flight operations with sufficient payload capacity.
o Navigation and Stability: Equipped with GPS modules and an advanced flight
controller (e.g., Pixhawk), the hexacopter supports autonomous navigation and
waypoint control. Additional sensors like LiDAR enhance obstacle detection and
collision avoidance.

2. AI-Powered Pest Identification System:

o High-Resolution Imaging: The drone integrates RGB and thermal cameras to
capture detailed aerial imagery, enabling precise detection of pests across varied
terrains and vegetation types.
o Generative AI Algorithms: A state-of-the-art Generative AI model processes
real-time image data to identify specific pest species. The model leverages a
curated dataset of pest images, ensuring high accuracy in pest classification and
behavior analysis.
o Onboard Computing: Devices like NVIDIA Jetson Nano facilitate real-time
image processing and decision-making directly on the drone, reducing latency
and dependency on external computing resources.

3. Ultrasonic Repellent System:

o Targeted Pest Repulsion: The system generates ultrasonic frequencies
specifically mapped to repel identified pest species without harming beneficial
organisms or the environment.
o Dynamic Frequency Adjustment: An Arduino-based function generator
dynamically adjusts frequencies based on AI-driven pest classification, ensuring
 targeted and efficient pest control.

4. Data Integration and Control:

o Real-Time Data Transmission: A telemetry module provides real-time data on
pest activity and drone operations to a ground station for monitoring and analysis.
o Autonomous Operation: The system utilizes preprogrammed flight paths for
efficient field coverage and includes manual override options for precision
interventions.

Scientific and Environmental Rationale:

• Eco-Friendly Approach: Unlike chemical pesticides, ultrasonic frequencies pose no harm

to humans, non-target organisms, or the environment. This aligns with sustainable
agricultural practices and global efforts to reduce chemical dependency.
• Precision Agriculture: The system enhances pest management accuracy by combining
real-time surveillance with targeted interventions, significantly minimizing resource
wastage and crop damage.
• Advanced Scalability: The modular design ensures adaptability for various crop types and
farming conditions, making the system suitable for small-scale farms and large agricultural
operations.

Anticipated Outcomes:

1. Pest Identification and Control: Accurate real-time identification and effective repulsion
of pests across diverse terrains and crop types.
2. Environmental Impact: A reduction in chemical pesticide usage, promoting healthier
ecosystems and safer food production.
3. Economic Benefits: Improved crop yields and reduced losses, offering a cost-effective
solution for farmers.
4. Scientific Advancements: Integration of AI and drone technology in agriculture,
contributing to the broader field of precision farming and smart agriculture.

The proposed hexacopter-based system represents a significant leap forward in sustainable

pest management, addressing critical challenges in modern agriculture with innovation,
efficiency, and environmental responsibility.

Development Plan (1.5 Years)

The development plan outlines a systematic approach to design, test, and implement the drone-
based pest management system over 18 months. The plan is aligned with the project's goals and the
challenge timeline.
Phase 1: Research and Design (Months 1–4)
• Conduct a literature review on pest-specific ultrasonic frequencies and AI-based pest
identification techniques.
• Finalize hardware components, including drone specifications, sensors, and ultrasonic
modules.
 • Develop a detailed system design, including payload integration and power management.
Phase 2: Prototype Development (Months 5–8)
• Assemble the hexacopter drone with integrated pest repellent modules.
• Develop and train Generative AI models using a curated dataset of pest images.
• Integrate the AI system with the drone's onboard computer for real-time pest identification.
Phase 3: Initial Testing and Iteration (Months 9–12)
• Conduct controlled environment tests to validate pest identification accuracy and ultrasonic
repellent efficacy.
• Optimize the drone's flight path, battery life, and payload stability.
• Incorporate feedback to improve system performance.
Phase 4: Field Deployment and Testing (Months 13–16)
• Deploy the drone in agricultural fields for real-world testing.
• Analyze the system's effectiveness in reducing pest activity and crop damage.
• Collect performance data to fine-tune AI algorithms and ultrasonic modules.
Phase 5: Final Validation and Documentation (Months 17–18)
• Perform extensive testing to ensure system reliability and scalability.
• Document the project's outcomes, including pest reduction statistics and environmental
impact.
• Prepare for large-scale implementation and potential commercialization.

Methodology: Future Model

Modules Used:
1. Hexacopter Platform:
- Drone frame (Hexacopter) with payload capacity.
- Brushless motors, ESCs, and Li-Po batteries for power and flight.
- GPS module and flight controller (e.g., Pixhawk).
- Obstacle detection sensors (Lidar or ultrasonic).

2. Camera and AI Integration:

- High-resolution camera (thermal/HD).
- Generative AI model trained for pest identification using datasets of pest images.
- Edge device (e.g., NVIDIA Jetson Nano) for onboard processing.

3. Ultrasonic Repellent System:

- Arduino-based frequency generator for real-time signal modulation.
- Ultrasonic transducers tuned to multiple frequencies.

Sub-Modules:
- Pest Frequency Mapping: A database linking pests to their specific ultrasonic
frequencies.
- Autonomous Navigation: GPS and waypoint control for predefined flight paths.
- Control and Communication: Ground station software for remote monitoring and
manual override.
Platform:
- Hardware: Hexacopter, Arduino/ESP32, NVIDIA Jetson Nano, high-resolution camera,
ultrasonic transducers.
- Software: Python-based AI modules, drone control software (e.g., Mission Planner).

Justifications:
1. Sustainability: Reduces chemical pesticide usage, aligning with global sustainable agriculture
goals.
2. Efficiency: Hexacopter mobility allows better coverage of tall crops and large fields compared
to ground-based systems.
3. Technological Edge: Combines AI-based precision pest identification with targeted ultrasonic
repellents.
4. Economic Impact: Protects crops from pests, increasing farmer yields and reducing losses.

Collaboration:
- AI Development: Collaboration with [AI Research Institution] for pest identification models.
- Drone Expertise: Partnership with [Drone Manufacturer/Consultant] for hexacopter design and
assembly.
- Agriculture Specialists: Working with local agricultural universities or organizations for field
tests and validation.

Expected Outcomes:
1. AI-Driven Pest Identification: Real-time recognition of pests using Generative
AI.
2. Precise Frequency Emission: Targeted ultrasonic frequencies mapped to specific
pests.
3. Aerial Coverage: Enhanced pest control across tall and large-scale crops.
4. Eco-Friendly Solution: Reduced reliance on chemical pesticides.
5. Scalable Model: Potential for commercialization in agricultural sectors globally.

Conclusion:
The proposed hexacopter-based precision pest management system represents a transformative
approach to addressing the critical challenges of modern agriculture. By integrating advanced
technologies such as Generative AI, ultrasonic frequency mapping, and drone mobility, this system
offers a groundbreaking alternative to traditional pest control methods that heavily rely on chemical
pesticides.

The innovation lies in its ability to achieve real-time pest identification and species-specific
repulsion while maintaining environmental sustainability. With its modular and scalable design,
the system is adaptable to diverse agricultural terrains and crop types, ensuring broad applicability
and significant impact.

The expected outcomes of this project go beyond immediate agricultural benefits. The reduction in
chemical pesticide usage directly contributes to healthier ecosystems, safer food production, and
alignment with global sustainability goals. Moreover, the technological advancements achieved
through this initiative—ranging from AI-driven pest recognition to drone-enabled surveillance—
position it as a benchmark in precision agriculture and smart farming.

In conclusion, this project exemplifies the fusion of science, technology, and sustainability to
address one of agriculture's most pressing challenges. Its successful implementation will not only
enhance crop yields and economic viability for farmers but also set a new standard for eco-friendly
agricultural practices worldwide. This initiative is poised to redefine pest management, making a
lasting contribution to the future of sustainable agriculture.

Riddhiraj Chanda

Ritam Das

Monodip Roy

Md Ayan Akhtar
Khan
Dr. Subhajit Das
Prof. (Dr.) Abir
Chattopadhyay
Detailed Budget (in INR):
Cost per Total Cost
Item Quantity Description/Description
Unit (INR) (INR)
Durable frame with
Hexacopter Frame 1 motor mounts and 10,500 10,500
landing gear
High-efficiency motors
Brushless Motors 6 1,500 9,000
(1000KV or similar)
Electronic Speed 30A ESCs compatible
6 1,600 9,600
Controllers (ESCs) with brushless motors
Carbon fiber or durable
Propellers 6 pairs 700 4,200
plastic propellers
6S 5000mAh high-
LiPo Battery 4 1500 6,000
capacity LiPo battery
Compatible LiPo battery
Battery Charger 1 charger with balancing 3000 3000
features
Advanced flight
Flight Controller 1 controller with GPS and 14,000 14,000
altitude hold features
GPS module for
GPS Module 1 2,500 2,500
autonomous navigation
High-definition camera
Camera Module 1 4,500 4,500
for pest identification
GenAI Integration Edge device or onboard
Hardware(RasBerry 1 processing unit for AI 10,000 10,000
Pi-5) computations
Transmitter and 6-channel RC transmitter
1 5,500 5,500
Receiver and receiver
Wireless telemetry
Telemetry Module 1 module for real-time 5,000 5,000
data transmission
Function generator with
Ultrasonic wave
ultrasonic frequency
Generator
1 output, custom circuit 5,000 5,000
circuit/modulle
made using passive
electronic components
Cables, connectors,
Miscellaneous
- screws, and mounting 3,000 3,000
Items
accessories
For AI training and
Laptop/ Computer 1 Development purposes 1,08,090 1,08,090
and system design
TOTAL : 1,99,690 INR