Skip to content

rriddhijain/RoadSense

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

RoadSense: Smart Infrastructure Monitoring at Scale

Every driver knows the pain. Potholes damage our vehicles, waste fuel, and cost the global economy billions in repairs annually. But here's the thing—road authorities don't know where they are either.

Introducing RoadSense: The world's first crowdsourced, hardware-agnostic road damage detection system that turns every vehicle into a sensor.

Using advanced accelerometer analysis and proprietary classification algorithms, RoadSense detects and classifies potholes and speed breakers in real-time—all from a simple embedded device no larger than a phone. No manual surveys. No guesswork. Just live, actionable intelligence.


Developed by:
Shivansh Srivastava · Riddhi Jain


What We Do

Real-Time Detection

  • Intelligent Hardware: Firmware running on ultra-low-power microcontrollers processes accelerometer data at 200Hz, automatically classifying road anomalies into severity levels
  • Three-Tier Classification System: Green (minor), Yellow (moderate), Red (severe) alerts—so municipal workers prioritize repairs strategically
  • Live Feedback: RGB LED indicators provide instant driver feedback while data streams to the cloud

Smart Dashboard & Heatmapping

  • Live Visualization: Watch incidents happen in real-time across your city
  • Heatmap Analytics: Geographic clustering reveals road infrastructure hotspots—the smoothness map for every street
  • Session Tracking: Analyze routes by vehicle, driver, or time period to identify repeat problem areas
  • Data Intelligence: Peak damage metrics, incident width, severity distribution, and trend analysis at your fingertips

The Technology Stack

  • Firmware: C-based embedded processing on TI MSPM0G3507 microcontroller
  • Host Platform: Python-powered dashboard with Plotly visualization
  • Data Pipeline: JSON-based event streaming with GPS integration (mock and real)
  • Edge Intelligence: Dual-channel ADC processing with noise filtering and baseline calibration

Why RoadSense Wins

Scalable: Deploy on any vehicle fleet—taxis, buses, delivery services, rideshare
Non-Intrusive: Works with existing CAN bus or UART communication
Real-Time: No cloud dependency—local processing with optional cloud sync
Data-Driven: Transforms raw accelerometer noise into actionable infrastructure intelligence
Cost-Effective: Leverages low-cost sensors already in vehicles


The Market Opportunity

  • Municipal Infrastructure: $500B+ annual road maintenance market globally
  • Fleet Operations: Insurance & logistics companies willing to pay for road condition intelligence
  • Smart City Integration: Governments mandating IoT-based infrastructure monitoring
  • Connected Vehicles: OEMs building in road intelligence as a premium feature

What You Get

A complete, deployable system ready to revolutionize how cities understand and maintain their roads. From hardware firmware to production-grade visualization—RoadSense is the missing layer in smart transportation infrastructure.

The roads are talking. We're listening.


Project Structure

firmware/        - Embedded C code for sensor processing & classification
host/            - Python dashboard for visualization & analytics
  pothole_dashboard.py  - Live Dash application with heatmaps
  pothole_logger.py     - Event logging system
  potholes.json         - Pothole event database
  speedbreakers.json    - Speed breaker event database

Getting Started

  1. Deploy Firmware: Flash the embedded code onto supported microcontrollers
  2. Run Logger: Start the logging service to collect accelerometer events
  3. Launch Dashboard: Access the live visualization dashboard on localhost
  4. Monitor & Act: Watch real-time road damage detection with severity-based prioritization

RoadSense: Where Infrastructure Meets Intelligence

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages

  • Python 63.6%
  • C 36.4%