Toolbox for solving CBFs in Python
Toolbox for implementing safety filters using Control Barrier Functions (CBFs) in python. Implemented functionality for CBFs and implicit CBFs that rely on a backup controller.
Implementation of the solver is provided for control-affine systems (with bounded input constraints) using cvxpy.
Defines Experiment class to efficiently run experiments with different (or none) CBFs
The toolbox is compatible with batched inputs (torch or tf) and individual inputs (numpy or jax)
Installation:
- Run
pip install -e .to install this project
Example available in examples/acc.ipynb
GitHub