Skip to content

QuantumCoderrr/AutoPulse

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

5 Commits
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

πŸš— AutoPulse AI

Redefining Automotive Reliability with Agentic AI

Python Streamlit Status

Team MotorCoderrrs | EY Techathon 6.0 Submission


πŸ“– Executive Summary

AutoPulse AI is an intelligent ecosystem that transforms vehicle maintenance from reactive to proactive. By leveraging Agentic AI, we enable vehicles to "think" and "act"β€”predicting mechanical failures before they happen, autonomously scheduling repairs with service centers, and providing critical Root Cause Analysis (RCA) data back to manufacturers to fix design defects.


🚩 Problem Statement

The current automotive maintenance lifecycle is broken:

  1. Vehicle Owners face unexpected breakdowns and safety hazards.
  2. Service Centers suffer from inefficient manual scheduling and uneven workload.
  3. OEMs (Manufacturers) lack real-time field data to identify and fix recurring design flaws.

πŸ’‘ Solution Overview

AutoPulse AI bridges these gaps using a multi-agent orchestration layer:

1. πŸš— For Vehicle Owners (Mobile App View)

Predictive Alerts: Real-time warnings for "Engine Overheating" or "Battery Failure" before breakdown occurs. Agentic Assistance: A Voice/Chat bot that answers queries ("What should I do?") and handles logistics. One-Click Booking: Autonomous negotiation with service centers for the best available slot.

2. πŸ› οΈ For Service Centers (Dashboard View)

Autonomous Inflow: Bookings appear automatically with pre-diagnosed issues. Parts Prediction: Inventory is allocated before the vehicle arrives, reducing turnaround time.

3. 🏭 For OEMs (Manufacturing View)

Closed-Loop Feedback: Field failure data is aggregated to identify recurring defects. Automated CAPA: The system generates Corrective Action/Preventive Action reports for R&D teams.


βš™οΈ Tech Stack

Prototype (Current Submission):

  • Language: Python 3.10
  • Frontend/UI: Streamlit (Simulating Mobile, Web, and Analytics Dashboards)
  • Visualization: Plotly & Pandas

Proposed Architecture (Full Scale): AI Framework: LangGraph / CrewAI (for multi-agent orchestration). ML Models: LSTM & Isolation Forest (for anomaly detection). Backend: FastAPI & PostgreSQL.


πŸš€ How to Run Locally

Prerequisites

  • Python 3.10 or higher installed.

Installation

  1. Clone the repository:

    git clone https://github.com/quantumcoderrr/autopulse.git
    cd autopulse
  2. Install dependencies:

    pip install -r requirements.txt
  3. Run the application:

    streamlit run autopulse_app.py

πŸ“Έ Screenshots / Prototype Video


πŸ‘₯ Team MotorCoderrrs

Name Role College
Sandip Ghosh System Architect & AI/ML Integration St. Thomas' College of Engineering & Technology
Sandhita Poddar UI/UX & Frontend Developer St. Thomas' College of Engineering & Technology
Abhirup Raha Backend & Data Integration St. Thomas' College of Engineering & Technology

About

πŸš— AutoPulse AI: An Agentic AI ecosystem for the automotive industry that predicts vehicle failures, autonomously schedules maintenance, and closes the feedback loop for manufacturers. Submissions for EY Techathon 6.0

Topics

Resources

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages