DisasterShield AI

DisasterShield AI is an advanced edge AI drone system designed to autonomously monitor environments, detect early signs of natural disasters, and provide real-time alerts for proactive disaster prevention. This project aims to leverage the power of AI, FPGA acceleration, and edge computing to create a robust and reliable system for disaster management and environmental monitoring.

Table of Contents

• Project Overview
• Features
• System Architecture
• Installation
• Usage
• Applications
• Contributing
• License
• Contact

Project Overview

DisasterShield AI integrates machine learning, environmental sensors, and autonomous drone technology to detect and respond to natural disasters like flash floods, droughts, and more. The system uses real-time data analysis powered by FPGA acceleration to identify potential threats and provides instant alerts to authorities, enabling timely interventions.

Features

• Autonomous Operation: Pre-programmed flight paths for autonomous monitoring of disaster-prone areas.
• Real-Time Detection: Edge AI models for detecting anomalies indicative of natural disasters.
• Multi-Modal Data Fusion: Combines image data with environmental sensor readings for enhanced accuracy.
• FPGA Acceleration: High-speed processing for real-time analysis and decision-making.
• Remote Alerts: Immediate notification to authorities through real-time data transmission.

System Architecture

The system comprises the following components:

1. Drone Platform: Equipped with cameras and environmental sensors for data collection.
2. Edge AI Processor: FPGA-based processing unit running AI models for real-time analysis.
3. Environmental Sensors: Humidity, temperature, air pressure, and soil moisture sensors for comprehensive environmental monitoring.
4. Data Transmission Module: For sending critical alerts and data to ground stations or cloud servers.

System Architecture Guidelines

Installation

Prerequisites

• Python 3.8+
• Xilinx PYNQ-Z2 Board
• Vitis AI
• Required Python libraries (listed in requirements.txt)

Setup Instructions

1. Clone the Repository:

   git clone https://github.com/4Kapture-Ai-Co/DisasterShield-AI.git
   cd DisasterShield-AI

2. Install Dependencies:

   pip install -r requirements.txt

3. Setup FPGA Environment:

   • Follow the instructions in the PYNQ documentation to set up your FPGA environment.

4. Deploy AI Models:

   • Use Vitis AI to deploy pre-trained models onto the FPGA.

Usage

Running the System

1. Power on the Drone and FPGA:

   • Ensure all sensors are connected and operational.

2. Initiate Flight Mission:

   • Pre-program the drone with desired flight paths and start the mission.

3. Monitor Real-Time Data:

   • The system will autonomously collect and analyze data. Alerts will be sent to the designated receivers upon detecting any anomalies.

4. Post-Flight Data Analysis:

   • Download and analyze collected data for further insights and model improvements.

Applications

• Disaster Prevention and Management: Early warning for floods, droughts, and other natural disasters.
• Environmental Monitoring: Track ecosystem health and contribute to climate change research.
• Agriculture: Precision farming with soil moisture monitoring and crop health analysis.
• Urban Planning: Flood risk mapping and water resource management.

Contributing

We welcome contributions to enhance DisasterShield AI. Please check the Contributing Guidelines for more details.

License

This project is licensed under the MIT License. See the LICENSE file for details.

Contact

For any inquiries or support, please contact Ryan Morales.