Have you ever tracked a loved one’s cab ride in real time, refreshing the screen every few minutes, heart racing every time the car takes a turn you didn’t expect? I’ve been there too — too many times.

It’s a familiar routine:

• Click on the live location link

• Zoom in

• Refresh

• Panic when the car turns slightly off the expected route

• Call: “All good?”

• Repeat

And it’s not just the person tracking — even the passenger often stays alert the entire ride, unable to rest or sleep, just in case something goes wrong.

This anxiety, though understandable, is exhausting. So I decided to build something to automate away that stress.

________________________________________

🧠 From Worry to Solution — A Route Deviation Detection System

I asked myself a simple question:

"What if we could detect when a cab deviates from its intended path — and alert the right people — without constant human monitoring?"

That question turned into a working prototype over a few weekends: A web-based Cab Safety Simulator powered by vector math, GPS logic, and OpenRouteService APIs.

________________________________________

🎬 What the Simulator Does

• I input a start and destination location

• The system fetches an optimal route using OpenRouteService

• I simulate the cab’s journey — including intentional off-route movements

• When the cab deviates from its expected vector path, the system triggers an alert

• A new route is fetched automatically, and tracking resumes

Think of it as a self-healing GPS guardian — no need for panic or manual intervention.

________________________________________

🛠️ How It’s Built — Under the Hood

Frontend:

• JavaScript + Leaflet.js for map visualization

• Smooth UI to draw routes and simulate car movement

Backend:

• Python backend for deviation detection

• Vector comparison logic determines if the cab's heading significantly diverges

• Alert and reroute trigger logic

External API:

• OpenRouteService (ORS) for real-time routing data

________________________________________

⚙️ Why Not Use Google or Apple’s Deviation Alerts?

Great question. Here’s the catch:

Google and Apple’s deviation alerts are designed for the person in the cab, not for someone monitoring remotely.

My system solves for a different use case:

✅ Works remotely — ideal for worried parents, partners, or friends

🧠 Direction-aware — uses vector math, not just turn deviations

🔁 Fully autonomous — reroutes and resumes tracking without human input

🔌 Scalable — embeddable in existing apps, ride APIs, or safety dashboards

🛰️ Expandable — useful for employee safety, fleet tracking, campus monitoring, etc.

________________________________________

🚀 What’s Already Built

✅ Real-time GPS tracking simulation

✅ Direction-aware deviation detection

✅ Autonomous rerouting and continued monitoring

✅ An interactive frontend simulator for testing

________________________________________

💭 What’s Next?

This started as a personal learning project, but it’s growing fast. Here’s what I’m exploring:

• 🌐 Field trials with actual GPS devices

• 📱 A mobile app prototype (Firebase or Supabase)

• 🔔 Real-time push alerts for deviation

• 🔒 Authentication and trip-sharing features