Rescue2U

AI-powered early disaster detection that saves lives through real-time alerts

Comment

Rescue2U

Protecting Homes. Empowering Communities.

Inspiration

Natural disasters such as floods, fires, and air pollution events often strike with little warning, especially in communities that lack affordable, real-time monitoring systems. In Malaysia and many other regions, delayed alerts can mean the difference between safety and loss of life or property.

Rescue2U was inspired by a simple but powerful idea:

Sensors don’t sleep — early warnings save time, and time saves lives.

We wanted to build a low-cost, AI-powered disaster detection system that not only detects danger early, but also connects communities, first responders, and donors on a single platform.

What it does

Rescue2U is a multi-hazard disaster detection and alert system that:

  • Collects real-time environmental data (temperature, humidity, smoke, rain, air quality, distance)
  • Uses AI models to predict disasters such as:
    • Floods
    • Fires
    • Air pollution events
  • Sends instant alerts to:
    • Safety responders (e.g., firefighters)
    • Residents in affected areas
  • Displays actionable insights on a web application
  • Enables community donations to support disaster victims

- Operates sustainably using solar power

How we built it

Hardware

  • Raspberry Pi (central controller)
  • Raspberry Pi Camera (visual monitoring)
  • Sensors:
    • Air quality sensor
    • Smoke sensor
    • Temperature & humidity sensor
    • Rain sensor
    • Ultrasonic sensor (flood level detection)
  • Solar panel for sustainable power

Software & AI

  • Real-time data collection on Raspberry Pi
  • AI models for disaster classification:
    • Flood
    • Fire
    • Air pollution
  • Web application for:
    • Live monitoring
    • Alerts
    • Public awareness
    • Donation management

System Flow

  1. Collect real-time sensor data
  2. Run AI prediction model
  3. Check: Is a disaster detected?
  4. If Yes → Send alerts to responders & residents
  5. If No → Continue monitoring

Challenges we ran into

  • Sensor calibration to avoid false alarms
  • Integrating multiple sensors with reliable timing
  • Balancing real-time performance with AI inference on limited hardware
  • Designing alerts that are fast, clear, and actionable
  • Ensuring the system remains low-cost and scalable

Each challenge pushed us to think carefully about optimization, reliability, and user impact.

Accomplishments that we’re proud of

  • Built a working multi-hazard detection system
  • Successfully integrated AI predictions with real-time sensors
  • Designed a dual-purpose platform:
    • Emergency alerts
    • Community donation support
  • Created a sustainable, solar-powered prototype
  • Delivered a solution with real social and environmental relevance

What we learned

  • AI is most powerful when combined with real-world data
  • Early detection is more valuable than complex post-disaster analysis
  • Hardware–software integration requires patience and testing
  • Designing for people under stress (during disasters) is as important as technical accuracy
  • Teamwork across roles (ML, UI/UX, Web, Systems) is critical for real-world impact

What’s next for Rescue2U

We plan to expand Rescue2U with:

  • AI-powered evacuation route planning
  • Gamified disaster preparedness to educate communities
  • Detection of additional disaster types
  • Wider deployment in high-risk regions
  • Improved AI accuracy with larger datasets

Rescue2U is more than a project — it’s a step toward safer, smarter, and more resilient communities.

Share this project:

Updates