Skip to content

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

228 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

AeonRoute: Graph-Augmented Generative EV Routing

AeonRoute is a sophisticated, hyper-intelligent routing engine designed specifically for Electric Vehicles (EVs) in complex urban environments. By merging Graph Neural Networks (GNNs) with Generative Adversarial Networks (GANs) and Reinforcement Learning (Q-Learning), AeonRoute doesn't just find the shortest path—it predicts traffic dynamics and optimizes for energy efficiency and charging infrastructure availability.

🚀 The Vision

In the transition to sustainable urban mobility, standard routing algorithms (like A* or Dijkstra) fall short. They ignore the non-linear energy consumption of EVs and the stochastic nature of city traffic. AeonRoute addresses this by:

  • Synthesizing Realistic Traffic: Using GANs to generate high-fidelity traffic patterns based on historical data.
  • Topological Intelligence: Leveraging GNNs to understand the relationship between road segments and charging nodes.
  • Adaptive Learning: Utilizing Q-Learning agents that "live" in the simulation, learning the most efficient routes through trial and error in a digital twin of Hyderabad.

🏗️ Architecture

AeonRoute is built as a modular microservices-oriented system:

🧠 Backend (Python / FastAPI)

  • Hybrid Routing Engine: Orchestrates multiple strategies, from classic graph search to neural-guided pathfinding.
  • Spatial Indexing: High-performance spatial lookups for charging stations and road segments.
  • Training Pipeline: Integrated services for GNN-GAN training and Q-Table updates.
  • Analytics Service: Real-time processing of simulation metrics and health data.

🎨 Frontend (React / Vite / Tailwind)

  • Dynamic Map Visualization: Real-time rendering of routes and station status.
  • Simulation Dashboard: Interactive controls for managing EV agents and viewing hardware-accelerated telemetry.
  • Training Monitor: Live SSE (Server-Sent Events) streams for tracking model convergence.

🛠️ Tech Stack

Category Technologies
Backend Python 3.10+, FastAPI, NetworkX, GeoPandas, Keras/TensorFlow
Frontend TypeScript, React, Vite, TailwindCSS, Framer Motion, Lucide React
Infrastructure Docker, Docker Compose, Make
Data GraphML (Hyderabad OpenStreetMap Data), JSON

🏁 Getting Started

Prerequisites

  • Python 3.10 or higher
  • Node.js 18+ & npm
  • Docker (optional, for containerized deployment)

Backend Setup

  1. Clone the repository:
    git clone https://github.com/yourusername/aeonroute.git
    cd aeonroute
  2. Create and activate a virtual environment:
    python -m venv .venv
    source .venv/bin/activate  # Windows: .venv\Scripts\activate
  3. Install dependencies:
    pip install -r requirements.txt
  4. Start the API:
    python backend_api.py

Frontend Setup

  1. Navigate to the frontend directory:
    cd frontend
  2. Install dependencies:
    npm install
  3. Start the development server:
    npm run dev

📊 Models and Training

AeonRoute utilizes three primary AI components located in src/models/:

  1. GNN-GAN: Generates synthetic but realistic traffic flows across the road graph.
  2. Q-Learning Agent: Optimizes route selection based on a reward function that balances time, energy, and battery safety.
  3. Spatial Index: Ensures sub-millisecond lookups for nearest charging stations.

To retrain the models:

python src/main.py --train

🧪 Testing

We maintain high standards for reliability. Run the test suite using:

# Backend tests
pytest tests/

# Frontend tests
cd frontend && npm test

📜 License

This project is licensed under the MIT License - see the LICENSE file for details.


Developed with a focus on sustainable urban infrastructure and next-generation AI.

SG-GAN-WITH-EV-ROUTING

About

AeonRoute is a full-stack EV navigation platform designed to reduce range anxiety and improve urban travel efficiency. It uses a hybrid routing engine that combines traditional pathfinding with deep learning and reinforcement learning to deliver smarter, energy-efficient routes in real time.

Topics

Resources

Contributing

Stars

2 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages