GlareGuard

Drive safely with GlareGuard; the smart navigation assistant that alerts you to when and where sun glare could impair your vision, keeping you prepared and in control.

Comment

Introduction

Drive safely with GlareGuard; the smart navigation assistant that alerts you to when and where sun glare could impair your vision, keeping you prepared and in control.

Inspiration

  • Driving into the sun isn’t just annoying; it’s dangerous. Sun glare is the second leading environmental cause of crashes.
  • Our goal was to build an end-to-end geospatial pipeline that predicts glare exposure along a planned route by fusing solar-geometry models, terrain DEMs, and urban obstruction data, all delivered in an interactive mobile UI.

What it does

  • Computes spatiotemporal glare risk along any driving route by sampling the path at sub-minute intervals.
  • Calculates solar azimuth & elevation at each sample point and evaluates relative heading of the vehicle to determine when drivers are facing the sun. -Performs line-of-sight obstruction tests using digital elevation models (SRTM) for terrain and OSM-derived building footprints + heights to determine whether the sun is blocked.
  • Produces a continuous glare score [0-1] per sample and aggregates to segment-level risk metrics (avg, max, duration-weighted).
  • Renders a color-coded polyline overlay on the map (green -> yellow -> red) and a time-series chart showing glare intensity over the trip.

How we built it

  • Frontend:
    • React Native (Expo) mobile app for iOS/Android with a MapView-based glare-risk overlay and a profile screen summarizing per-segment metrics.
    • Handles route selection, live requests to the backend, and visualization of returned glare data.
  • Routing:
    • Google Maps Directions API for trip geometry, distances, durations, and polyline decoding.

  • Backend:

    • Python 3 + FastAPI serving REST endpoints.
    • PySolar for precise solar angle calculations and ephemeris data.
    • OSMnx for building geometries + heights.
    • SRTM elevation data for terrain (mountains, hills).

  • Overview of our glare_score function, to calculate azimuth and elevation values for sun placement, and delta and head values for relation to vehicle position and direction:

    def glare_score(lat, lon, when, heading_deg):
az, el = _solar(lat, lon, when)
elev = _elev_term(el)
delta = _angdiff(az, heading_deg)
head = _heading_term(delta)
    • Custom geospatial pipeline that samples points along routes and runs line-of-sight checks to the sun.

  • Security: All API keys stored server-side, never exposed to the client.

Challenges we ran into

  • API key security → solved by using a server proxy for all external requests.
  • Geospatial math → had to align different coordinate systems (lat/lon vs projected meters).
  • Incomplete OSM building data → added fallback estimates for building heights.
  • Performance issues → optimized by adaptive sampling (denser near sunrise/sunset).

Accomplishments that we're proud of

  • Built a scientifically accurate glare prediction model that combines solar geometry, terrain, and urban data.
  • Validated predictions against known glare trouble spots.
  • Designed a secure architecture that prevents key leaks.
  • Created a mobile app that makes complex geospatial modeling usable for everyday drivers.

What we learned

  • Deep dive into coordinate reference systems and geospatial data handling.
  • Learned app development using Expo
  • How to combine astronomy libraries with mapping APIs.
  • Practical techniques for securing mobile-to-server API communication.
  • That even simple user experiences often require bridging multiple technical domains.

What's next for GlareGuard

  • Native integration with Google Maps turn-by-turn navigation for real-time glare warnings.
  • Push notifications / calendar hooks to warn users about upcoming high-glare trips.
  • Crowdsourced feedback to refine obstruction models.
  • Long-term vision: AR windshield overlays showing glare risk in real time.
Share this project:

Updates