Skip to content

[NeurIPS 2021] Implementation of Probabilistic Barrier Certificates submitted to NeurIPS 2021 safeRL workshop

License

Notifications You must be signed in to change notification settings

athindran/ProBF

Repository files navigation

ProBF

Implementation of ProBF: Learning Probabilistic Safety Certificates with Barrier Functions presented at NeurIPS 2021 SafeRL workshop. (https://arxiv.org/abs/2112.12210)

This code compares the our ProBF-GP framework with the prior art learned Control Barrier Functions which uses a neural network (LCBF-NN) [2]. We build upon the code in [1]. We add a controller that uses the mean and variance of the predictions to solve the ProBF-convex program.

image

This page will be updated further

Authors

Athindran Ramesh Kumar

Sulin Liu

Jaime F. Fisac

Ryan P. Adams

Peter J. Ramadge

Please reach out for any questions!

References

[1] Python simulation and hardware library for learning and control. https://github.com/learning-and-control/core.git. Accessed: 2021-11-17.

[2] A. Taylor, A. Singletary, Y. Yue, and A. Ames. Learning for safety-critical control with control barrier functions. In Learning for Dynamics and Control, pages 708–717. PMLR, 2020.

About

[NeurIPS 2021] Implementation of Probabilistic Barrier Certificates submitted to NeurIPS 2021 safeRL workshop

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages