** AIDR-Net: AI-Driven Disaster Response Network**
Inspiration
Natural disasters and emergency situations often damage communication infrastructure, making it difficult for affected people to report incidents and receive help. We were inspired by the challenge of creating a disaster response system that remains functional even when internet connectivity is unavailable. Our goal was to build a solution that combines AI, offline communication, and real-time reporting to support both civilians and emergency responders during critical situations.
What it does
AI-Driven Disaster Response Network (AIDR-Net) is a mobile application that enables users to report emergencies through a one-tap SOS feature. The system automatically classifies incidents such as fires, floods, accidents, and medical emergencies using AI. Reports are stored locally, displayed on an interactive map, and shared with nearby devices using offline communication technologies. The application also provides alerts and notifications about nearby incidents, ensuring users remain informed even in low-connectivity environments.
How we built it
We developed AIDR-Net using Flutter and Dart to create a cross-platform mobile application. SQLite was used for local data storage, enabling offline functionality. We implemented a custom machine learning classifier to categorize disaster reports automatically. Google Maps integration was used for visualizing incident locations, while Bluetooth and Nearby Connections technologies enabled device-to-device communication without internet access. Flutter Local Notifications were integrated to deliver emergency alerts to users.
Challenges we ran into
One of the biggest challenges was implementing reliable communication in offline environments. Ensuring that emergency reports could be shared between devices without internet connectivity required careful integration of Bluetooth-based communication. Another challenge was balancing AI classification accuracy with mobile performance constraints. We also faced difficulties managing location services, permissions, and efficient storage of disaster reports while maintaining a smooth user experience.
Accomplishments that we're proud of
We successfully built a fully functional offline-first disaster response prototype that combines AI-powered classification, local storage, map visualization, and peer-to-peer communication. The ability to report and share emergency information without internet connectivity is a major achievement. We are also proud of creating a scalable architecture that can be extended with cloud services, advanced AI models, and real-time disaster monitoring systems in the future.
What we learned
Through this project, we gained hands-on experience in Flutter development, SQLite database management, machine learning integration, map-based visualization, and offline communication technologies. We also learned the importance of designing resilient systems that can operate under challenging real-world conditions. Additionally, we improved our understanding of disaster management workflows and how technology can support emergency response efforts.
What's next for AI-Driven Disaster Response Network
Our future plans include integrating Firebase for cloud synchronization, enhancing the AI module with advanced NLP models, adding real-time weather and satellite data integration, and developing an administrative dashboard for disaster analytics. We also aim to implement mesh-network communication to expand offline coverage and create predictive analytics features that help authorities identify and respond to potential disaster risks more effectively.
Log in or sign up for Devpost to join the conversation.