NeuralExplorer: State Space Exploration of Closed Loop Control Systems Using Neural Networks

Manish Goyal, Parasara Sridhar Duggirala
Proceedings of the 2nd Conference on Learning for Dynamics and Control, PMLR 120:697-697, 2020.

Abstract

In this paper, we propose a framework for performing state space exploration of closed loop control systems. For closed loop control systems, we introduce the notion of inverse sensitivity function and present a mechanism for approximating inverse sensitivity by a neural network. This neural network can be used for generating trajectories that reach a destination (or a neighborhood around it). We demonstrate the effectiveness of our approach by applying it to standard nonlinear dynamical systems, nonlinear hybrid systems, and also neural network based feedback control systems.

Cite this Paper

BibTeX
@InProceedings{pmlr-v120-goyal20a, title = {NeuralExplorer: State Space Exploration of Closed Loop Control Systems Using Neural Networks}, author = {Goyal, Manish and Duggirala, Parasara Sridhar}, booktitle = {Proceedings of the 2nd Conference on Learning for Dynamics and Control}, pages = {697--697}, 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/goyal20a/goyal20a.pdf}, url = {https://proceedings.mlr.press/v120/goyal20a.html}, abstract = {In this paper, we propose a framework for performing state space exploration of closed loop control systems. For closed loop control systems, we introduce the notion of inverse sensitivity function and present a mechanism for approximating inverse sensitivity by a neural network. This neural network can be used for generating trajectories that reach a destination (or a neighborhood around it). We demonstrate the effectiveness of our approach by applying it to standard nonlinear dynamical systems, nonlinear hybrid systems, and also neural network based feedback control systems.} }
Endnote
%0 Conference Paper %T NeuralExplorer: State Space Exploration of Closed Loop Control Systems Using Neural Networks %A Manish Goyal %A Parasara Sridhar Duggirala %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-goyal20a %I PMLR %P 697--697 %U https://proceedings.mlr.press/v120/goyal20a.html %V 120 %X In this paper, we propose a framework for performing state space exploration of closed loop control systems. For closed loop control systems, we introduce the notion of inverse sensitivity function and present a mechanism for approximating inverse sensitivity by a neural network. This neural network can be used for generating trajectories that reach a destination (or a neighborhood around it). We demonstrate the effectiveness of our approach by applying it to standard nonlinear dynamical systems, nonlinear hybrid systems, and also neural network based feedback control systems.
APA
Goyal, M. & Duggirala, P.S.. (2020). NeuralExplorer: State Space Exploration of Closed Loop Control Systems Using Neural Networks. Proceedings of the 2nd Conference on Learning for Dynamics and Control, in Proceedings of Machine Learning Research 120:697-697 Available from https://proceedings.mlr.press/v120/goyal20a.html.

Related Material