/star-sailor

Solar Sail Dynamics Simulator

Primary LanguagePythonMIT LicenseMIT

Star Sailor - Solar Sail Dynamics Simulator

This is a project to kick off my 4th year undergraduate thesis. The goal of this simulator is for me to develop an intuition for working with solar sails and to develop a tool that can be used to simulate the dynamics of a solar sail in a variety of scenarios. The code in here will be used later one to help me develop a steering law for trajectory optimization.

Getting Started

Star Sailor runs on PyPy, which speeds up regular Python code. I am using this to compare against my previous projects built on regular Python, ctypes, and cffi. However, the code should be drop-in compatible with regular Python 3.8+.

To install, simply run pip install git+https://github.com/itchono/star-sailor. This will install the package and all of its dependencies. To run the simulator, simply run python -m star-sailor.

If installing on PyPy, use PyPy 3.8, and add the flag --extra-index-url https://antocuni.github.io/pypy-wheels/manylinux2010/ to the pip install command.

Dynamics

Star sailor is a 2D simulator using Cowell's method of propagation with the following accelerations:

Solar Radiation Pressure

$$a_{\text{SRP}} = \left[2 P_{\text{Sun}} \cdot \frac{A}{m} \cdot \cos^2 \alpha \right] \hat{n}$$ where $P_{\text{Sun}}$ is the solar irradiance at 1 AU, $A$ is the area of the sail, $m$ is the mass of the sail, and $\alpha$ is the angle between the sail normal ($\hat{n}$) and the sun vector.

This is for an ideal sail, which is perfectly reflective.

Gravity

$$a_{\text{grav}} = -\frac{\mu}{r^3} \boldsymbol{r}$$ where $\mu$ is the gravitational parameter of the central body, and $\vec{r}$ is the position vector of the spacecraft relative to the central body.

This is for a spherical central body (i.e. without J2 effects).

Integration

Time integration is done using the Dormand-Prince 8(7) method with a variable step size. The error tolerance is set to $10^{-12}$. The integration is done using the SciPy library. However, I am also looking at making my own integrator using extrapolation methods.