A Python library for estimating the flight dynamics of gliding aircraft.
To simplify its use in common engineering applications (such as control
modeling and statistical filtering), the glider dynamics are encoded as
differential equations (ẋ = f(x, u)).
- Source code: https://github.com/pfheatwole/glidersim/
- License: permissive (MIT)
- Documentation: https://pfheatwole.github.io/glidersim/
- Questions: for general questions, please use the discussion forum.
- Support: to report a problem, please use the issue tracker.
- Novel wing geometry model optimized for nonlinear designs
- Paraglider component models parametrized by basic technical data
- Paraglider system models that accept all three primary control inputs: brakes, accelerator, and weight shift (direct manipulation of the risers and individual lines is not supported)
- Stateless models that are easy understand, use, extend, and test
- Nonlinear aerodynamics using a fast nonlinear lifting-line theory model
- Graceful accuracy degradation near stall (important for paragliders, which frequently operate at relatively high angles of attack)
- Supports non-uniform wind fields and non-uniform wind vectors along the aerodynamic surfaces, enabling simulations involving local wind sheer, thermal lift and sink, wing rotation, etc.
- Includes paraglider system dynamics models that account for apparent mass (nonlinear dynamics due to the motion of an object through a fluid that become more significant for low-density volumes such as parafoils)
- Flight dynamics are encoded as numerical state derivatives, providing a simple model interface
- Includes a rudimentary simulator for generating flight trajectories from predefined wing control and wind field inputs
- Open source with a permissive license, built using the Python scientific computing stack (NumPy, SciPy, and Numba). Usable from within embedded Python interpreters, such as the Python consoles in Blender and FreeCAD.
Equally important is what it does not do:
- The library is not suitable for studying behaviors involving turbulence, wing deformations, rapid wing maneuvers, or post-stall behaviors.
- The library is not a structural simulator. The models are not adequate for designing or testing the structural integrity of a design; they assume rigid-body dynamics without estimating internal forces. (For example, it neglects suspension line tension, cell billowing, canopy wrinkling, etc.)
- The library is not focused on engineering control input sequences or wind field models. The bundled simulator is for executing simulations given some inputs, not for designing them.
- The library is not focused on execution time. Although speed is important, the design prioritizes readability, functionality, and flexibility.
To install the latest release of the package into a Python virtual environment:
$ pip install pfh.glidersimOr, for the latest development version:
$ pip install git+https://github.com/pfheatwole/glidersim- Design overview: for the background and architecture of the library, refer to the design guide.
- Examples: to use the library to build glider models or run flight simulations, refer to the usage guide.
- API: to program using the library, refer to the library reference.
- Derivations: for a deeper discussion of the modeling choices, mathematical derivations, and literature references, refer to my thesis: Parametric Paraglider Modeling.
This software has been stable for my purposes, but should be considered "alpha" quality. The design (including the API) needs more users to test it before it could be considered stable. Also the estimates produced by the nonlinear aerodynamics method needs more verification.