Geometric constraint solver for Python. Uses numerical optimisation (via SciPy) to determine the solution to a given set of length and angular constraints.
See the documentation for more details.
As stated in the documentation, this project is just a bit of fun. If the developer or anyone else willing to contribute has time, here are some ideas for improvements to the library:
- Implement overconstraint detection (e.g. using techniques presented https://doi.org/10.1007/s11831-020-09509-y).
- Implement symbolic parameters such that constraints can reference parameters of primitives. For example, it would be nice to be able to constrain a line to match the length of another line, even if the length of that other line is a free parameter.
- Implement symbolic solving, which will make solutions more robust and probably easier to find when the initial conditions are far from the optimal. SolveSpace for example implements this.
- Implement GUI to let users build and solve a system visually.
- Implement solution selection. Some problems (such as the [https://github.com/SeanDS/pygeosolve/blob/master/examples/triangle.py](triangle example)) have multiple valid solutions to the given constraints. Right now the solution chosen is the one the optimiser happens to converge on, which is sometimes not the one that's closest to the initial conditions. It is usually best choose the solution that's closest to the initial conditions in order to capture the user's intentions. This probably requires a different solver since the current one provides only one solution.
Sean Leavey https://github.com/SeanDS/