Metadata-Version: 2.4
Name: quadrotor_koopman_datafree
Version: 0.1.0
Summary: Koopman-based data-free lifting for quadrotor dynamics on SE(3)
Home-page: https://github.com/yourusername/quadrotor_koopman_datafree
Author: Santosh Rajkumar
License: MIT
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.8
Description-Content-Type: text/markdown
Requires-Dist: numpy
Requires-Dist: scipy
Requires-Dist: casadi
Requires-Dist: matplotlib

# quadrotor_koopman_datafree

A lightweight Python package for **Koopman-based data-free lifting of quadrotor dynamics on SE(3)**. This package analytically constructs lifted linear representations of quadrotor dynamics without requiring any trajectory data or machine learning. It supports both **NumPy** (numerical) and **CasADi** (symbolic) versions of the Koopman lifting maps for control and Model Predictive Control (MPC) applications.

---

## ✨ Key Features

- ✅ Koopman state lifting on **SE(3)** for quadrotors  
- ✅ Analytical (data-free) Koopman operators  
- ✅ Compatible with **linear MPC, LQR, and optimal control**  
- ✅ Supports **CasADi symbolic modeling**  
- ✅ Includes utilities for state conversion, noise, rotation matrices  
- ✅ Maps between actual and lifted control spaces (`u ↔ U`, `u_tilde`)  
- ✅ Clean and modular implementation  

---

## 📦 Installation

From PyPI:
```bash
pip install quadrotor_koopman_datafree
