Skip to content

csmdg7/Hump_Detection_App_Showcase

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 

Repository files navigation

🚦 Road Safety Intelligence System (Hump Detection App)

🔒 PATENT PENDING & PRIVATE REPOSITORY Please Note: The source code for this project is currently maintained in a private repository due to ongoing intellectual property protection and publication in the Indian Patent Office (IPO) Journal. This showcase repository documents the system architecture, mathematical methodologies, and data pipelines utilized in the project.

📖 Project Overview

Road traffic injuries in densely populated areas are frequently exacerbated by unmarked or poorly visible traffic calming measures (speed humps). Traditional detection systems rely on expensive vehicle hardware or continuous cloud connectivity.

This project introduces a sensor-independent, mobile-first edge computing architecture built in Flutter. By combining continuous GPS telemetry, mathematical proximity calculations, and local offline data persistence, the system provides real-time audio-visual hazard alerts to drivers without requiring internet access or vehicle retrofitting.


⚙️ Core System Architecture

[ Mobile GPS Sensor ] ──(Live Coordinates)──> [ Haversine Proximity Engine ]
                                                         │
[ OpenStreetMap (OSM) Nodes ] ──(Pre-loaded)──> [ SQLite Local Database ] 
                                                         │
[ Text-to-Speech (TTS) ] <──(Proximity Alert)────────────┘

1. Geospatial Tracking & Telemetry

Instead of relying on hardware accelerometers which suffer from mechanical vibration noise across different vehicle types, this system extracts high-frequency GPS streams directly from the mobile device. This approach guarantees hardware independence and immediate scalability across any consumer smartphone.

2. The Haversine Proximity Engine

The core computational engine utilizes the Haversine formula to calculate the great-circle distance between the user's live coordinates and the cached coordinates of known road humps. This allows the system to deterministically trigger alerts when the proximity threshold is breached.

3. Edge Computing & Offline Persistence (SQLite)

Because cellular data drops in rural areas or tunnels, the application operates on a strict Offline-First paradigm.

  • Local Database: Geospatial nodes and hazard coordinates are stored natively on the device using SQLite.
  • Zero Cloud Dependency: The proximity engine executes entirely on the local device (Edge Computing), ensuring zero-latency alerts and functionality even in dead zones.

4. Multi-Modal Accessibility Alerts

To minimize driver distraction, the application bypasses standard UI visual checks by utilizing native Text-to-Speech (TTS). When the SQLite database confirms an upcoming hazard within the Haversine radius, the system forces a hands-free auditory warning.


🛠️ Technology Stack & Deployment

  • Frontend Framework: Flutter (Dart) deployed as a local Android APK.
  • Geospatial Data: OpenStreetMap (OSM) coordinate structures.
  • Local Persistence Layer: SQLite (Offline relational data caching).
  • Mathematics: Applied spherical trigonometry (Haversine Formula).
  • Hardware Integration: Native GPS / Location Services, Text-to-Speech (TTS) engines.
  • Deployment: Executed locally as an Edge-Computing application to ensure real-time latency limits.

👨‍💻 Research & Development Team

  • Lead Developer / System Architect: Chetana Srinivasa Murthy
  • Development Team: Amrutha Shindhe C, Chandana, Haripriya KP
  • Project Guides: Dr. Shantakumar B Patil (Professor and Associate Dean of Research), Mr. Manoj Chunangat (Co-Guide)
  • Institution: Sai Vidya Institute of Technology, Department of Computer Science and Engineering

🏫 Target Environment & Validation Pipeline

📍 Localized Campus Deployment

The system was initially conceptualized and engineered as a localized pilot for our college campus. Navigating institutional roadways safely requires hyper-local geographic data that mainstream mapping services (like Google Maps) often fail to track or update accurately.

👥 Human-in-the-Loop Crowd Sourcing

To ensure scalability, the architecture includes a decentralized user-reporting module allowing drivers to log new or unmarked road humps dynamically.

To prevent data pollution (vandalism, duplicate entries, or inaccurate GPS drift coordinates), the system utilizes an Administrative Verification Gateway:

  1. User Submission: Drivers upload raw coordinate streams via the client application.
  2. Staging Queue: Data points are held in a separate staging table inside the architecture.
  3. Manual Validation: System administrators manually verify the coordinates against physical campus routes before promoting the nodes to the master SQLite production database, ensuring absolute data integrity.

About

An offline-first, edge-computing road safety app built for campus navigation. Uses continuous GPS telemetry and the Haversine formula to detect road hazards, featuring a crowd-sourced coordinate reporting system with administrative validation gateways. [Patent-Pending]

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors