14. Solving ordinary differential equations#
We are developing a library for solving nonlinear ordinary differential equations. First we build an abstraction of nonlinear functions, and implement solvers for nonlinear equations. On this basis we can simply define time-stepping algorithms. We apply it to multi-body systems. OpenGL visualization is provided based on the python interface to threejs.
Solving non-linear equations
Explicit and implicit time-stepping methods
Connecting masses with springs
Introducing constraints
Fork that github repo ASC-ODE, and clone it via
git clone https://github.com/my_fork_on_github/ASC-ODE.git
It contains the basic linear algebra package nanoblas as an git submodule. Populate it using
cd ASC-ODE
git submodule update --init
You are welcome to replace nanoblas by your preferred linear algebra package.
ASC-ODE has the following dependencies which must be installed:
pybind11
lapack
Then you can use cmake to build ASC-ODE:
mkdir build
cd build
cmake ..
make