Ballistix

Advanced Trajectory Prediction & Simulation Platform

A full-stack application combining physics-based simulation with machine learning
for accurate projectile trajectory prediction and real-time visualization.

Live Demo • API Docs • Features • Quick Start • API Reference

---

Overview

Ballistix is an advanced platform for simulating and predicting projectile and missile trajectories, combining physics-based simulation with machine learning and a modern, interactive UI.

• Modern UI: Dark theme, animated hero, glassmorphism, responsive design
• Navigation: Footer and CTA buttons on the About page now switch views (Home, Ballistic, Analytics, Settings, Launch Simulator)
• Physics Engine: Runge-Kutta 4 integration with ISA atmospheric model
• Machine Learning: Random Forest model with 98% R² accuracy
• Real-time Visualization: Animated trajectory plots, multiple projectile types
• Missile Presets: 20+ real-world missile profiles

---

Features

Feature	Description
Modern UI	Dark theme, animated hero, glassmorphism, responsive
Footer Navigation	Home, Ballistic, Analytics, Settings (About page)
Launch Simulator	CTA button launches trajectory simulation
Physics Simulation	Newtonian mechanics, altitude-dependent gravity, air density
ML Predictions	Random Forest, instant impact predictions
Analytics Dashboard	Compare physics vs ML, track accuracy
Missile Database	ICBMs, SLBMs, cruise missiles, real specs
3D Globe View	Visualize long-range trajectories
User Sessions	Per-user history and settings

---

Quick Start

Prerequisites

Node.js >= 18.0.0
Python >= 3.10
npm >= 9.0.0

Option 1: Use Live Version

Frontend: https://ballistix.vercel.app
API Docs: https://ballistix.onrender.com/docs

Option 2: Run Locally

# Clone repository
git clone https://github.com/rana-rohit/projectile-prediction.git
cd projectile-prediction

# Backend setup
cd backend
python -m venv venv
venv\Scripts\activate        # Windows
# source venv/bin/activate   # macOS/Linux
pip install -r requirements.txt
uvicorn app.main:app --reload --port 8000

# Frontend setup (new terminal)
cd frontend
npm install
npm run dev

Access:

• Frontend: http://localhost:5173
• API: http://localhost:8000
• API Docs: http://localhost:8000/docs

---

Project Structure

projectile-prediction/
├── backend/
│   ├── app/
│   │   ├── main.py            # FastAPI API endpoints
│   │   ├── sim.py             # Physics engine (RK4 + ISA)
│   │   ├── model_store.py     # ML model management
│   │   ├── database.py        # SQLAlchemy config
│   │   └── db_models.py       # Data models
│   ├── models/
│   │   ├── rf_model.joblib    # Trained Random Forest
│   │   ├── scaler.joblib      # Feature scaler
│   │   └── model_metadata.json
│   ├── requirements.txt
│   └── runtime.txt
│   └── static/
│       └── ballistic/         # Static previews
│
├── frontend/
│   ├── src/
│   │   ├── app.jsx            # Main app & routing
│   │   ├── Ballistic.jsx      # Missile simulation
│   │   ├── analytics.jsx      # Analytics dashboard
│   │   ├── About.jsx          # Modern landing page (with navigation)
│   │   ├── Settings.jsx       # User preferences
│   │   ├── Auth.jsx           # Authentication
│   │   ├── AnimationCanvas.jsx
│   │   └── styles.css         # Global styles
│   ├── index.html
│   ├── vercel.json
│   └── package.json
│
├── DEPLOYMENT.md
├── LICENSE
└── README.md

---

Physics Engine

The simulation uses Runge-Kutta 4 (RK4) numerical integration with:

Equations of Motion

dx/dt = vₓ
dy/dt = vᵧ
dvₓ/dt = -k·(ρ/ρ₀)·|v|·vₓ
dvᵧ/dt = -g(h) - k·(ρ/ρ₀)·|v|·vᵧ

Atmospheric Model (ISA)

Altitude	Layer	Model
0-11 km	Troposphere	Linear temp decrease
11-20 km	Lower Stratosphere	Isothermal (216.65K)
20-32 km	Upper Stratosphere	Linear temp increase
32-84 km	Mesosphere	Exponential decay
>84 km	Thermosphere	Negligible density

Physical Constants

R_EARTH = 6,371,000 m      # Earth radius
G = 6.67430×10⁻¹¹          # Gravitational constant
M_EARTH = 5.972×10²⁴ kg    # Earth mass
ρ₀ = 1.225 kg/m³           # Sea-level air density

---

API Reference

Base URL

Production: https://ballistix.onrender.com
Local: http://localhost:8000

Endpoints

Health Check

GET /health

{
  "status": "ok",
  "model_loaded": true,
  "api_version": "1.0.0"
}

Predict Trajectory

POST /predict
Content-Type: application/json

{
  "v0": 300,
  "angle": 45,
  "drag": 0.01,
  "dt": 0.01,
  "release_height": 0
}

Response:

{
  "xs": [0.0, 2.1, 4.2, ...],
  "ys": [0.0, 1.5, 2.9, ...],
  "impact_physics": 9174.32,
  "impact_ml": 9168.45,
  "max_height": 2295.68,
  "max_range": 9174.32,
  "flight_time": 61.24,
  "trajectory_points": 6124
}

Model Info

GET /model/info

Retrain Model

POST /retrain
Content-Type: application/json

{
  "n_samples": 1200,
  "use_forest": true
}

cURL Examples

# Health check
curl https://ballistix.onrender.com/health

# Run prediction
curl -X POST https://ballistix.onrender.com/predict \
  -H "Content-Type: application/json" \
  -d '{"v0":300,"angle":45,"drag":0.01,"dt":0.01,"release_height":0}'

---

Machine Learning

Model Details

Parameter	Value
Algorithm	Random Forest Regressor
Trees	100 estimators
R² Score	~0.99
Training Samples	1,200
Features	v0, angle, drag, release_height
Target	Impact distance

Training Pipeline

Physics Simulation → Generate Dataset → Train/Test Split (80/20)
                                              ↓
                                        StandardScaler
                                              ↓
                                      Random Forest Fit
                                              ↓
                                    Cross-Validation (5-fold)
                                              ↓
                                      Save Model + Metadata

---

Missile Database

Pre-configured profiles with verified specifications:

ICBMs

Missile	Country	Range	Speed
Minuteman III	🇺🇸 USA	13,000 km	Mach 23
Trident II D5	🇺🇸 USA	11,300 km	Mach 24
RS-28 Sarmat	🇷🇺 Russia	18,000 km	Mach 20.7
DF-41	🇨🇳 China	12,000-15,000 km	Mach 25
Agni-V	🇮🇳 India	5,500-8,000 km	Mach 24

SLBMs

Missile	Country	Range	Platform
Trident II	🇺🇸 USA	11,300 km	Ohio-class
R-30 Bulava	🇷🇺 Russia	8,300 km	Borei-class
JL-3	🇨🇳 China	10,000+ km	Type 096

Cruise Missiles

Missile	Country	Range	Speed
Tomahawk	🇺🇸 USA	2,500 km	Mach 0.75
BrahMos	🇮🇳/🇷🇺	450 km	Mach 2.8
Kalibr	🇷🇺 Russia	2,500 km	Mach 0.8

Sources: CSIS Missile Threat, FAS, Jane's Defence

---

Tech Stack

Frontend

• React 18 - UI framework
• Vite 5 - Build tool
• Recharts - Data visualization
• Three.js - 3D globe
• Lucide React - Icons
• Axios - HTTP client
• Modern CSS - Glassmorphism, gradients, responsive

Backend

• FastAPI - Web framework
• Uvicorn - ASGI server
• Scikit-learn - ML models
• NumPy/SciPy - Numerical computing
• Pandas - Data processing
• SQLAlchemy - ORM
• Pydantic - Data validation

Deployment

• Vercel - Frontend hosting
• Render - Backend hosting
• GitHub - Version control

---

Navigation & Usage

• The About page features a modern landing design with animated hero, glassmorphism, and a dark theme.
• Footer navigation buttons (Home, Ballistic, Analytics, Settings) and the "Launch Simulator" CTA are fully functional, switching views using the onNavigate prop.
• Navigation view names: trajectory, ballistic, analytics, settings.

---

---

Performance

Metric	Value
API Response (predict)	< 100ms
Physics Simulation	~100k steps/sec
ML Inference	< 5ms
Frontend Bundle	645 KB (gzipped: 183 KB)
Lighthouse Score	85+

---

Security

• ✅ Input validation with Pydantic
• ✅ CORS configured per environment
• ✅ No sensitive data in client storage
• ✅ HTTPS enforced in production
• ⚠️ Rate limiting recommended for production

---

Contributing

1. Fork the repository
2. Create feature branch (git checkout -b feature/amazing-feature)
3. Commit changes (git commit -m 'Add amazing feature')
4. Push to branch (git push origin feature/amazing-feature)
5. Open a Pull Request

---

License

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

MIT License
Copyright (c) 2025 Rohit Rana

---

🌟 Star this repo if you found it useful!

Live Demo • API Docs • Report Issues