Sentinel-X9 PRO: Edge-AI Driver Drowsiness Interceptor

A serverless in-browser Edge-AI system using real-time facial mesh and EAR telemetry to intercept driver micro-sleep with instant acoustic sirens and visual strobes.

Comment

Inspiration

What it does

How we built it

Challenges we ran into

Accomplishments that we're proud of

What we learned

What's next for Sentinel-X9 PRO: Edge-AI Driver Drowsiness Interceptor

Inspiration

Road accidents caused by driver fatigue, micro-sleep, and late-night highway exhaustion claim thousands of lives every year. Traditional hardware countermeasure systems (infrared ocular glasses, specialized automotive rigs) are costly and inaccessible for regular vehicle owners and transport fleets. I built Sentinel-X9 PRO to prove that life-saving computer vision can run on any driver's existing smartphone browser without specialized external hardware.

What it does

Sentinel-X9 PRO is an autonomous, in-browser Edge-AI system designed for real-time driver drowsiness mitigation:

  • Facial Mesh & Eye Tracking: Uses MediaPipe to track 468+ facial landmarks and map precise eye contours in real time.
  • Metric Computation: Continuously calculates the Eye Aspect Ratio (EAR) and driver head orientation.
  • Sub-Second Interception: If eyelids remain closed past the temporal threshold, the system triggers a multi-frequency synthetic alarm alongside an active visual strobe.
  • Telemetry Graph: Renders real-time drowsiness risk indicators and logs event instances without transmitting raw video to any external server.

How I built it

  • Computer Vision Pipeline: Built with MediaPipe Face Mesh running client-side through WebGL/WebAssembly acceleration.
  • Acoustic Engine: Integrated using the native HTML5 Web Audio API with sawtooth oscillators to generate immediate sound without external audio assets.
  • Frontend & HUD: Built using modern HTML5, CSS3, and Vanilla JavaScript with HTML5 Canvas for real-time contour and telemetry plotting.
  • Deployment: Hosted directly on GitHub Pages for zero-latency, cross-platform mobile browser execution.

Challenges I ran into

  • Calibrating the Eye Aspect Ratio (EAR) thresholds to eliminate false positives caused by natural eye blinks versus actual micro-sleep.
  • Ensuring zero lag and reliable real-time tracking on mobile browsers with varying hardware capabilities.
  • Managing browser autoplay policies to reliably arm and trigger the Web Audio siren on sudden drowsy events.

Accomplishments that I'm proud of

  • Created a fully operational, zero-cost Edge-AI countermeasure system that runs completely client-side.
  • Successfully built a solution addressing real-world highway safety without requiring external servers or subscription hardware.

What's next for Sentinel-X9 PRO

  • Direct integration with vehicle CAN-bus/OBD-II hardware to trigger engine deceleration or seat vibration motors.
  • Multi-camera fleet dashboard support for commercial transport operators.

Built With

Share this project:

Updates

Submission history