Advanced AI-powered racing intelligence platform for Toyota GR "Hack the Track" Hackathon
ORIS is a complete real-time racing analytics system that combines 5 specialized AI models with actual Toyota GR Cup telemetry data to provide strategic racing insights. Built for the Real-Time Analytics category with full hackathon compliance.
Real-Time Analytics Category
Uses Toyota GR Cup Race Data from COTA, Road America, Sebring, VIR, Barber, Sonoma
Live Processing of telemetry streams with AI predictions
Interactive Dashboard with strategy recommendations
Professional Racing Interface built for Toyota drivers
ORIS combines five specialized AI models with real Toyota GR Cup telemetry data to deliver comprehensive racing intelligence:
- Real-time telemetry analysis from actual Toyota races (60Hz data processing)
- AI-powered strategy recommendations using 5 specialist models
- Live database integration with InfluxDB 3 Core for data correlation
- Real-time 2D track visualization of racing lines and track analysis
- Complete settings management with user access control
- MINERVA (~800K params): Strategic racing intelligence with multi-head attention for pit strategy and race tactics
- ATLAS (~1.2M params): Spatial track intelligence with position-aware attention and track memory systems
- IRIS (~1M params): Vehicle dynamics intelligence with specialized throttle/brake/balance analysis
- CHRONOS (~1.3M params): Timing intelligence with LSTM networks and temporal pattern recognition
- PROMETHEUS (~1.4M params): Predictive analytics with multi-horizon forecasting and race outcome modeling
- Live Strategy Advisor: Real-time Standard/Push/Conserve mode recommendations
- Toyota Data Integration: Processing actual GR Cup race telemetry files
- Live Data Feed: Real-time telemetry streaming from field cars with toggle control
- Database Management: InfluxDB 3 Core with user access control and live connection monitoring
- Interactive Settings: Complete configuration interface for database, AI models, alerts
- Responsive Design: Optimized for all screen sizes and racing environments
- Barber Motorsports Park
- Circuit of the Americas (COTA)
- Road America
- Sebring International Raceway
- Sonoma Raceway
- Virginia International Raceway (VIR)
- React 19.1 with TypeScript 5.9
- Vite 7.1 for lightning-fast development
- Custom 2D Canvas for track visualization
- Recharts for real-time telemetry graphs
- Lucide React for professional racing icons
- Responsive CSS with mobile optimization
- FastAPI REST API serving AI model predictions
- PyTorch 2.2+ for neural network inference
- 5 Specialized AI Models (MINERVA, ATLAS, IRIS, CHRONOS, PROMETHEUS)
- Concurrent server startup with automatic dependency management
- Flexible Training: Models trainable locally, Google Colab, or cloud platforms
- InfluxDB 3 Core OSS for telemetry time-series data
- Toyota GR Cup Data from 6 professional racing circuits
- CSV/JSON processing for historical race analysis
- Real-time data correlation and vector storage
- Node.js 18+ and npm
- Python 3.12+
- Git for cloning the repository
- Clone the repository:
git clone https://github.com/AutomataControls/ToyotaGR.git
cd ToyotaGR- Install all dependencies (Node.js + Python):
npm install- Start the application (auto-starts both servers):
npm run devThat's it! The command will:
- Create Python virtual environment automatically
- Install PyTorch, FastAPI, and all Python dependencies
- Start AI API server on port 8000
- Start React app on port 5173
- Run both servers concurrently
# Install Python dependencies only
npm run setup:venv
# Start AI API server only
npm run start:api
# Start React frontend only
npm run start:frontend
# Production build
npm run buildFor full database functionality:
- Install InfluxDB 3 Core:
# Download from InfluxDB website
# Or use Docker: docker run -p 8181:8181 influxdb:3.0- Start InfluxDB server:
influxdb3 serve --node-id host01 --object-store file --data-dir ~/.influxdb3- Configure in ORIS Settings:
- URL:
http://localhost:8181 - Username:
your_username - Token:
your_token - Database:
toyota_gr_telemetry
- URL:
ToyotaGR/
├── src/
│ ├── components/ # React components
│ │ ├── layout/ # Header, Sidebar, Layout
│ │ ├── common/ # Reusable UI components
│ │ ├── dashboard/ # Main dashboard widgets
│ │ ├── strategy/ # Strategy advisor components
│ │ └── telemetry/ # Telemetry visualization
│ ├── pages/ # Main application pages
│ │ ├── Dashboard.tsx # Main racing dashboard
│ │ ├── Strategy.tsx # Strategy recommendations
│ │ ├── Telemetry.tsx # Live telemetry graphs
│ │ ├── Database.tsx # Live data feed management
│ │ ├── Help.tsx # Documentation and AI model details
│ │ └── Settings.tsx # System configuration
│ ├── services/ # API clients and data services
│ │ ├── aiModelAPI.py # FastAPI server for AI models
│ │ ├── aiModelClient.ts # TypeScript AI client
│ │ ├── liveDataService.ts # Live data feed management
│ │ └── toyotaDataLoader.ts # Toyota data processing
│ ├── models/ # AI model implementations
│ │ ├── minerva/ # Strategic analysis model
│ │ ├── atlas/ # Spatial reasoning model
│ │ ├── iris/ # Vehicle dynamics model
│ │ ├── chronos/ # Timing analysis model
│ │ └── prometheus/ # Predictive modeling
│ └── data/ # Toyota GR Cup race data
│ └── tracks/ # COTA, Road America, etc.
├── venv/ # Python virtual environment (auto-created)
├── dist/ # Production build output
├── requirements.txt # Python dependencies
└── package.json # Node.js dependencies & scripts
# Get strategy recommendations from MINERVA
POST http://localhost:8000/predict/minerva
# Get spatial analysis from ATLAS
POST http://localhost:8000/predict/atlas
# Get vehicle dynamics from IRIS
POST http://localhost:8000/predict/iris
# Get timing predictions from CHRONOS
POST http://localhost:8000/predict/chronos
# Get future predictions from PROMETHEUS
POST http://localhost:8000/predict/prometheus
# Get all model status
GET http://localhost:8000/models/status# Get live data status
GET http://localhost:8000/live-data/status
# Toggle live data feed
POST http://localhost:8000/live-data/toggle
# Get connected field cars
GET http://localhost:8000/live-data/cars
# Get recent telemetry packets
GET http://localhost:8000/live-data/recent?limit=20
# Field car data connection endpoint
POST http://localhost:8000/live-data/connect- 5 AI Specialists: Each model provides specialized racing insights
- Real Toyota Data: Processing actual GR Cup telemetry from 6 tracks
- Live Data Feed: Real-time telemetry streaming from field cars with toggle control
- Live Updates: Real-time strategy recommendations and connection monitoring
- Database Integration: InfluxDB 3 Core for time-series data with live feed management
- Full Settings: User management, database config, model status, live data configuration
- Comprehensive Help: Detailed AI model architecture documentation
- Responsive UI: Works on all devices and screen sizes
- Auto-Deploy: One command starts everything
- Startup Time: < 30 seconds for full system (both servers + AI models)
- AI Inference: < 100ms per model, < 500ms for all 5 models
- React Build: ~30 seconds optimized production bundle
- Memory Usage: ~1GB total (React + Python + AI models)
- Real-time Processing: 60Hz telemetry data streams supported
# Development with hot reload
npm run dev
# Build production bundle
npm run build
# Preview production build
npm run preview
# Python environment setup
npm run setup:venv
# Run linting
npm run lint🏆 COMPETING IN: Real-Time Analytics Category
Why ORIS Wins:
- Real AI Implementation: 5+ million parameter PyTorch neural networks (not mock data)
- Live Race Engineer Tool: Professional real-time decision support system
- Toyota Integration: Processes actual GR Cup telemetry with exact parameter specifications
- Production Ready: Complete full-stack implementation with live data capabilities
Toyota Requirements Met:
- Real Toyota GR Cup data integration
- Professional racing interface
- Real-time analytics processing
- Strategic insights for drivers
- Scalable architecture
MIT License
🏁 Built for Toyota GR "Hack the Track" Hackathon 2024
Category: Real-Time Analytics
Team: AutomataControls
- Toyota Racing Development for GR Cup telemetry data access
- InfluxDB for time-series database technology
- FastAPI for high-performance Python web framework
- React/Vite for modern frontend development
Ready to race? Run npm run dev and experience the future of racing intelligence!