Probabilistic Safety Constraints for Learned High Relative Degree System Dynamics

Mohammad Javad Khojasteh, Vikas Dhiman, Massimo Franceschetti, Nikolay Atanasov
Proceedings of the 2nd Conference on Learning for Dynamics and Control, PMLR 120:781-792, 2020.

Abstract

This paper focuses on learning a model of system dynamics online while satisfying safety constraints. Our motivation is to avoid offline system identification or hand-specified dynamics models and allow a system to safely and autonomously estimate and adapt its own model during online operation. Given streaming observations of the system state, we use Bayesian learning to obtain a distribution over the system dynamics. In turn, the distribution is used to optimize the system behavior and ensure safety with high probability, by specifying a chance constraint over a control barrier function.

Cite this Paper

BibTeX
@InProceedings{pmlr-v120-khojasteh20a, title = {Probabilistic Safety Constraints for Learned High Relative Degree System Dynamics}, author = {Khojasteh, Mohammad Javad and Dhiman, Vikas and Franceschetti, Massimo and Atanasov, Nikolay}, booktitle = {Proceedings of the 2nd Conference on Learning for Dynamics and Control}, pages = {781--792}, year = {2020}, editor = {Bayen, Alexandre M. and Jadbabaie, Ali and Pappas, George and Parrilo, Pablo A. and Recht, Benjamin and Tomlin, Claire and Zeilinger, Melanie}, volume = {120}, series = {Proceedings of Machine Learning Research}, month = {10--11 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v120/khojasteh20a/khojasteh20a.pdf}, url = {https://proceedings.mlr.press/v120/khojasteh20a.html}, abstract = {This paper focuses on learning a model of system dynamics online while satisfying safety constraints. Our motivation is to avoid offline system identification or hand-specified dynamics models and allow a system to safely and autonomously estimate and adapt its own model during online operation. Given streaming observations of the system state, we use Bayesian learning to obtain a distribution over the system dynamics. In turn, the distribution is used to optimize the system behavior and ensure safety with high probability, by specifying a chance constraint over a control barrier function.} }
Endnote
%0 Conference Paper %T Probabilistic Safety Constraints for Learned High Relative Degree System Dynamics %A Mohammad Javad Khojasteh %A Vikas Dhiman %A Massimo Franceschetti %A Nikolay Atanasov %B Proceedings of the 2nd Conference on Learning for Dynamics and Control %C Proceedings of Machine Learning Research %D 2020 %E Alexandre M. Bayen %E Ali Jadbabaie %E George Pappas %E Pablo A. Parrilo %E Benjamin Recht %E Claire Tomlin %E Melanie Zeilinger %F pmlr-v120-khojasteh20a %I PMLR %P 781--792 %U https://proceedings.mlr.press/v120/khojasteh20a.html %V 120 %X This paper focuses on learning a model of system dynamics online while satisfying safety constraints. Our motivation is to avoid offline system identification or hand-specified dynamics models and allow a system to safely and autonomously estimate and adapt its own model during online operation. Given streaming observations of the system state, we use Bayesian learning to obtain a distribution over the system dynamics. In turn, the distribution is used to optimize the system behavior and ensure safety with high probability, by specifying a chance constraint over a control barrier function.
APA
Khojasteh, M.J., Dhiman, V., Franceschetti, M. & Atanasov, N.. (2020). Probabilistic Safety Constraints for Learned High Relative Degree System Dynamics. Proceedings of the 2nd Conference on Learning for Dynamics and Control, in Proceedings of Machine Learning Research 120:781-792 Available from https://proceedings.mlr.press/v120/khojasteh20a.html.

Related Material