From raw data to safety: Reducing conservatism by set expansion

Mohammad Bajelani, Klaske Van Heusden
Proceedings of the 6th Annual Learning for Dynamics & Control Conference, PMLR 242:1305-1317, 2024.

Abstract

In response to safety concerns associated with learning-based algorithms, safety filters have been proposed as a modular technique. Generally, these filters heavily rely on the system’s model, which is contradictory if they are intended to enhance a data-driven or end-to-end learning solution. This paper extends our previous work, a purely Data-Driven Safety Filter (DDSF) based on Willems’ lemma, to an extremely short-sighted and non-conservative solution. Specifically, we propose online and offline sample-based methods to expand the safe set of DDSF and reduce its conservatism. Since this method is defined in an input-output framework, it can systematically handle both unknown and time-delay LTI systems using only one single batch of data. To evaluate its performance, we apply the proposed method to a time-delay system under various settings. The simulation results validate the effectiveness of the set expansion algorithm in generating a notably large input-output safe set, resulting in safety filters that are not conservative, even with an extremely short prediction horizon.

Cite this Paper

BibTeX
@InProceedings{pmlr-v242-bajelani24a, title = {From raw data to safety: {R}educing conservatism by set expansion}, author = {Bajelani, Mohammad and Heusden, Klaske Van}, booktitle = {Proceedings of the 6th Annual Learning for Dynamics & Control Conference}, pages = {1305--1317}, year = {2024}, editor = {Abate, Alessandro and Cannon, Mark and Margellos, Kostas and Papachristodoulou, Antonis}, volume = {242}, series = {Proceedings of Machine Learning Research}, month = {15--17 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v242/bajelani24a/bajelani24a.pdf}, url = {https://proceedings.mlr.press/v242/bajelani24a.html}, abstract = {In response to safety concerns associated with learning-based algorithms, safety filters have been proposed as a modular technique. Generally, these filters heavily rely on the system’s model, which is contradictory if they are intended to enhance a data-driven or end-to-end learning solution. This paper extends our previous work, a purely Data-Driven Safety Filter (DDSF) based on Willems’ lemma, to an extremely short-sighted and non-conservative solution. Specifically, we propose online and offline sample-based methods to expand the safe set of DDSF and reduce its conservatism. Since this method is defined in an input-output framework, it can systematically handle both unknown and time-delay LTI systems using only one single batch of data. To evaluate its performance, we apply the proposed method to a time-delay system under various settings. The simulation results validate the effectiveness of the set expansion algorithm in generating a notably large input-output safe set, resulting in safety filters that are not conservative, even with an extremely short prediction horizon.} }
Endnote
%0 Conference Paper %T From raw data to safety: Reducing conservatism by set expansion %A Mohammad Bajelani %A Klaske Van Heusden %B Proceedings of the 6th Annual Learning for Dynamics & Control Conference %C Proceedings of Machine Learning Research %D 2024 %E Alessandro Abate %E Mark Cannon %E Kostas Margellos %E Antonis Papachristodoulou %F pmlr-v242-bajelani24a %I PMLR %P 1305--1317 %U https://proceedings.mlr.press/v242/bajelani24a.html %V 242 %X In response to safety concerns associated with learning-based algorithms, safety filters have been proposed as a modular technique. Generally, these filters heavily rely on the system’s model, which is contradictory if they are intended to enhance a data-driven or end-to-end learning solution. This paper extends our previous work, a purely Data-Driven Safety Filter (DDSF) based on Willems’ lemma, to an extremely short-sighted and non-conservative solution. Specifically, we propose online and offline sample-based methods to expand the safe set of DDSF and reduce its conservatism. Since this method is defined in an input-output framework, it can systematically handle both unknown and time-delay LTI systems using only one single batch of data. To evaluate its performance, we apply the proposed method to a time-delay system under various settings. The simulation results validate the effectiveness of the set expansion algorithm in generating a notably large input-output safe set, resulting in safety filters that are not conservative, even with an extremely short prediction horizon.
APA
Bajelani, M. & Heusden, K.V.. (2024). From raw data to safety: Reducing conservatism by set expansion. Proceedings of the 6th Annual Learning for Dynamics & Control Conference, in Proceedings of Machine Learning Research 242:1305-1317 Available from https://proceedings.mlr.press/v242/bajelani24a.html.

Related Material