Safe Nonlinear Control Using Robust Neural Lyapunov-Barrier Functions

Charles Dawson, Zengyi Qin, Sicun Gao, Chuchu Fan
Proceedings of the 5th Conference on Robot Learning, PMLR 164:1724-1735, 2022.

Abstract

Safety and stability are common requirements for robotic control systems; however, designing safe, stable controllers remains difficult for nonlinear and uncertain models. We develop a model-based learning approach to synthesize robust feedback controllers with safety and stability guarantees. We take inspiration from robust convex optimization and Lyapunov theory to define robust control Lyapunov barrier functions that generalize despite model uncertainty. We demonstrate our approach in simulation on problems including car trajectory tracking, nonlinear control with obstacle avoidance, satellite rendezvous with safety constraints, and flight control with a learned ground effect model. Simulation results show that our approach yields controllers that match or exceed the capabilities of robust MPC while reducing computational costs by an order of magnitude. We provide source code at github.com/dawsonc/neural_clbf/.

Cite this Paper

BibTeX
@InProceedings{pmlr-v164-dawson22a, title = {Safe Nonlinear Control Using Robust Neural Lyapunov-Barrier Functions}, author = {Dawson, Charles and Qin, Zengyi and Gao, Sicun and Fan, Chuchu}, booktitle = {Proceedings of the 5th Conference on Robot Learning}, pages = {1724--1735}, year = {2022}, editor = {Faust, Aleksandra and Hsu, David and Neumann, Gerhard}, volume = {164}, series = {Proceedings of Machine Learning Research}, month = {08--11 Nov}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v164/dawson22a/dawson22a.pdf}, url = {https://proceedings.mlr.press/v164/dawson22a.html}, abstract = {Safety and stability are common requirements for robotic control systems; however, designing safe, stable controllers remains difficult for nonlinear and uncertain models. We develop a model-based learning approach to synthesize robust feedback controllers with safety and stability guarantees. We take inspiration from robust convex optimization and Lyapunov theory to define robust control Lyapunov barrier functions that generalize despite model uncertainty. We demonstrate our approach in simulation on problems including car trajectory tracking, nonlinear control with obstacle avoidance, satellite rendezvous with safety constraints, and flight control with a learned ground effect model. Simulation results show that our approach yields controllers that match or exceed the capabilities of robust MPC while reducing computational costs by an order of magnitude. We provide source code at github.com/dawsonc/neural_clbf/.} }
Endnote
%0 Conference Paper %T Safe Nonlinear Control Using Robust Neural Lyapunov-Barrier Functions %A Charles Dawson %A Zengyi Qin %A Sicun Gao %A Chuchu Fan %B Proceedings of the 5th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2022 %E Aleksandra Faust %E David Hsu %E Gerhard Neumann %F pmlr-v164-dawson22a %I PMLR %P 1724--1735 %U https://proceedings.mlr.press/v164/dawson22a.html %V 164 %X Safety and stability are common requirements for robotic control systems; however, designing safe, stable controllers remains difficult for nonlinear and uncertain models. We develop a model-based learning approach to synthesize robust feedback controllers with safety and stability guarantees. We take inspiration from robust convex optimization and Lyapunov theory to define robust control Lyapunov barrier functions that generalize despite model uncertainty. We demonstrate our approach in simulation on problems including car trajectory tracking, nonlinear control with obstacle avoidance, satellite rendezvous with safety constraints, and flight control with a learned ground effect model. Simulation results show that our approach yields controllers that match or exceed the capabilities of robust MPC while reducing computational costs by an order of magnitude. We provide source code at github.com/dawsonc/neural_clbf/.
APA
Dawson, C., Qin, Z., Gao, S. & Fan, C.. (2022). Safe Nonlinear Control Using Robust Neural Lyapunov-Barrier Functions. Proceedings of the 5th Conference on Robot Learning, in Proceedings of Machine Learning Research 164:1724-1735 Available from https://proceedings.mlr.press/v164/dawson22a.html.

Related Material