TY - GEN
T1 - Worst-case Satisfaction of STL Specifications Using Feedforward Neural Network Controllers
T2 - 2020 Information Theory and Applications Workshop, ITA 2020
AU - Yaghoubi, Shakiba
AU - Fainekos, Georgios
N1 - Funding Information:
This work was partially supported by the NSF awards CNS 1350420, IIP-1361926, and the NSF I/UCRC Center for Embedded Systems.
Funding Information:
This work was partially supported by the NSF awards CNS 1350420, IIP-1361926, and the NSF I/UCRC Center for Embedded Systems. This article appears as part of the ESWEEK-TECS special issue and was presented at the International Conference on Embedded Software (EMSOFT) 2019. Authors’ addresses: Shakiba Yaghoubi, School of Computing, Informatics, and Decision Systems Engineering, Arizona State University, Tempe, Arizona, 85281, syaghoub@asu.edu; Georgios Fainekos, School of Computing, Informatics, and Decision Systems Engineering, Arizona State University, Tempe, Arizona, 85281, fainekos@asu.edu. Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from permissions@acm.org. © 2019 Association for Computing Machinery. 1539-9087/2019/9-ART127 $15.00 https://doi.org/10.1145/
Publisher Copyright:
© 2020 IEEE.
PY - 2020/2/2
Y1 - 2020/2/2
N2 - In this paper, a reinforcement learning approach for designing feedback neural network controllers for nonlinear systems is proposed. Given a Signal Temporal Logic (STL) specification which needs to be satisfied by the system over a set of initial conditions, the neural network parameters are tuned in order to maximize the satisfaction of the STL formula. The framework is based on a max-min formulation of the robustness of the STL formula. The maximization is solved through a Lagrange multipliers method, while the minimization corresponds to a falsification problem. We present our results on a vehicle and a quadrotor model and demonstrate that our approach reduces the training time more than 50 percent compared to the baseline approach.
AB - In this paper, a reinforcement learning approach for designing feedback neural network controllers for nonlinear systems is proposed. Given a Signal Temporal Logic (STL) specification which needs to be satisfied by the system over a set of initial conditions, the neural network parameters are tuned in order to maximize the satisfaction of the STL formula. The framework is based on a max-min formulation of the robustness of the STL formula. The maximization is solved through a Lagrange multipliers method, while the minimization corresponds to a falsification problem. We present our results on a vehicle and a quadrotor model and demonstrate that our approach reduces the training time more than 50 percent compared to the baseline approach.
KW - Reinforcement Learning
KW - Signal Temporal Logic
KW - neural network controller
KW - • Computer systems organization ? Robotic control
KW - • Theory of computation ? Adversarial learning
UR - http://www.scopus.com/inward/record.url?scp=85097350477&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85097350477&partnerID=8YFLogxK
U2 - 10.1109/ITA50056.2020.9244969
DO - 10.1109/ITA50056.2020.9244969
M3 - Conference contribution
AN - SCOPUS:85097350477
T3 - 2020 Information Theory and Applications Workshop, ITA 2020
BT - 2020 Information Theory and Applications Workshop, ITA 2020
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 2 February 2020 through 7 February 2020
ER -