System-level safety guard: Safe tracking control through uncertain neural network dynamics models

Xiao Li, Yutong Li, Anouck Girard, Ilya Kolmanovsky
Proceedings of the 6th Annual Learning for Dynamics & Control Conference, PMLR 242:968-979, 2024.

Abstract

The Neural Network (NN), as a black-box function approximator, has been considered in many control and robotics applications. However, difficulties in verifying the overall system safety in the presence of uncertainties hinder the deployment of NN modules in safety-critical systems. In this paper, we leverage the NNs as predictive models for trajectory tracking of unknown dynamical systems. We consider controller design in the presence of both intrinsic uncertainty and uncertainties from other system modules. In this setting, we formulate the constrained trajectory tracking problem and show that it can be solved using Mixed-integer Linear Programming (MILP). The proposed MILP-based approach is empirically demonstrated in robot navigation and obstacle avoidance through simulations. The demonstration videos are available at https://xiaolisean.github.io/publication/2023-11-01-L4DC2024.

Cite this Paper

BibTeX
@InProceedings{pmlr-v242-li24a, title = {System-level safety guard: {S}afe tracking control through uncertain neural network dynamics models}, author = {Li, Xiao and Li, Yutong and Girard, Anouck and Kolmanovsky, Ilya}, booktitle = {Proceedings of the 6th Annual Learning for Dynamics & Control Conference}, pages = {968--979}, 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/li24a/li24a.pdf}, url = {https://proceedings.mlr.press/v242/li24a.html}, abstract = {The Neural Network (NN), as a black-box function approximator, has been considered in many control and robotics applications. However, difficulties in verifying the overall system safety in the presence of uncertainties hinder the deployment of NN modules in safety-critical systems. In this paper, we leverage the NNs as predictive models for trajectory tracking of unknown dynamical systems. We consider controller design in the presence of both intrinsic uncertainty and uncertainties from other system modules. In this setting, we formulate the constrained trajectory tracking problem and show that it can be solved using Mixed-integer Linear Programming (MILP). The proposed MILP-based approach is empirically demonstrated in robot navigation and obstacle avoidance through simulations. The demonstration videos are available at https://xiaolisean.github.io/publication/2023-11-01-L4DC2024.} }
Endnote
%0 Conference Paper %T System-level safety guard: Safe tracking control through uncertain neural network dynamics models %A Xiao Li %A Yutong Li %A Anouck Girard %A Ilya Kolmanovsky %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-li24a %I PMLR %P 968--979 %U https://proceedings.mlr.press/v242/li24a.html %V 242 %X The Neural Network (NN), as a black-box function approximator, has been considered in many control and robotics applications. However, difficulties in verifying the overall system safety in the presence of uncertainties hinder the deployment of NN modules in safety-critical systems. In this paper, we leverage the NNs as predictive models for trajectory tracking of unknown dynamical systems. We consider controller design in the presence of both intrinsic uncertainty and uncertainties from other system modules. In this setting, we formulate the constrained trajectory tracking problem and show that it can be solved using Mixed-integer Linear Programming (MILP). The proposed MILP-based approach is empirically demonstrated in robot navigation and obstacle avoidance through simulations. The demonstration videos are available at https://xiaolisean.github.io/publication/2023-11-01-L4DC2024.
APA
Li, X., Li, Y., Girard, A. & Kolmanovsky, I.. (2024). System-level safety guard: Safe tracking control through uncertain neural network dynamics models. Proceedings of the 6th Annual Learning for Dynamics & Control Conference, in Proceedings of Machine Learning Research 242:968-979 Available from https://proceedings.mlr.press/v242/li24a.html.

Related Material