Timezone: »
We present neural differentiable predictive control (DPC) method for learning constrained neural control policies for uncertain linear systems. DPC is formulated as a differentiable problem whose computational graph architecture is inspired by classical model predictive control (MPC) structure. In particular, the optimization of the neural control policy is based on automatic differentiation of the MPC loss function through a differentiable closed-loop system dynamics model. We show that DPC can learn constrained neural control policies to stabilize systems with unstable dynamics, track time-varying references, and satisfy state and input constraints without the prior need of a supervisory MPC controller.
Author Information
Jan Drgona (Pacific Northwest National Laboratory)
I am a data scientist in the Physics and Computational Sciences Division (PCSD) at Pacific Northwest National Laboratory, Richland, WA. My current research interests fall in the intersection of model-based optimal control, constrained optimization, and machine learning.
Aaron Tuor (Pacific Northwest National Laboratory)
My undergraduate background is in Linguistics, Philosophy, and Mathematics. I was actively engaged in Machine Learning research, with the Hutchinson Research Group at WWU from the beginning of my graduate studies. I have extensively studied state of the art recommendation systems. Most recently I have been studying deep learning approaches for several domains including language processing, image processing, and games. My current main research focus is on detecting anomaly in network traffic logs to aid in filtering information for cyber security analysts.
Draguna Vrabie (Pacific Northwest National Laboratory)
Draguna Vrabie is a chief data scientist at Pacific Northwest National Laboratory, in the Data Sciences and Machine Intelligence group where she serves as team leader for the Autonomous Learning and Reasoning team. Her work is at the intersection of control system theory and machine learning and is aimed at the design of adaptive decision and control systems. Her current focus is on deep learning methodologies and algorithms for design and operation of high-performance, cyber-physical systems. Prior to joining PNNL in 2015, she was a senior scientist at United Technologies Research Center, in East Hartford, Connecticut. She published two books on optimal control and reinforcement learning, and more than fifty journal and conference papers, with more than five thousand citations. She has chapters in the Control Systems Handbook and in the Handbook on Computational Intelligence. She has served on the editorial board of IEEE Transactions on Control Systems Technology and IEEE Transactions of Neural Networks and Learning Systems, as a program committee member for international symposia on control, computing, and machine learning, and as a technical reviewer for conferences, journals and government funding programs. Vrabie holds a doctorate in electrical engineering from the University of Texas at Arlington, and an ME and BE in automatic control and computer engineering from Gheorghe Asachi Technical University, Iaşi, Romania.
More from the Same Authors
-
2020 : Physics-constrained Deep Recurrent Neural Models of Building Thermal Dynamics »
Jan Drgona -
2023 Competition: The CityLearn Challenge 2023 »
Zoltan Nagy · Kingsley Nweye · Sharada Mohanty · Ruchi Choudhary · Max Langtry · Gregor Henze · Jan Drgona · Sourav Dey · Alfonso Capozzoli · Mohamed Ouf -
2022 Workshop: Tackling Climate Change with Machine Learning »
Peetak Mitra · Maria João Sousa · Mark Roth · Jan Drgona · Emma Strubell · Yoshua Bengio -
2022 Competition: The CityLearn Challenge 2022 »
Zoltan Nagy · Kingsley Nweye · Sharada Mohanty · Siva Sankaranarayanan · Jan Drgona · Tianzhen Hong · Sourav Dey · Gregor Henze -
2021 Workshop: Tackling Climate Change with Machine Learning »
Maria João Sousa · Hari Prasanna Das · Sally Simone Fobi · Jan Drgona · Tegan Maharaj · Yoshua Bengio -
2021 : Differentiable Parametric Optimization Approach to Power System Load Modeling »
Jan Drgona · Andrew August · Elliott Skomski -
2021 Poster: On the Stochastic Stability of Deep Markov Models »
Jan Drgona · Sayak Mukherjee · Jiaxin Zhang · Frank Liu · Mahantesh Halappanavar -
2019 : Lunch + Poster Session »
Frederik Gerzer · Bill Yang Cai · Pieter-Jan Hoedt · Kelly Kochanski · Soo Kyung Kim · Yunsung Lee · Sunghyun Park · Sharon Zhou · Martin Gauch · Jonathan Wilson · Joyjit Chatterjee · Shamindra Shrotriya · Dimitri Papadimitriou · Christian Schön · Valentina Zantedeschi · Gabriella Baasch · Willem Waegeman · Gautier Cosne · Dara Farrell · Brendan Lucier · Letif Mones · Caleb Robinson · Tafara Chitsiga · Victor Kristof · Hari Prasanna Das · Yimeng Min · Alexandra Puchko · Alexandra Luccioni · Kyle Story · Jason Hickey · Yue Hu · Björn Lütjens · Zhecheng Wang · Renzhi Jing · Genevieve Flaspohler · Jingfan Wang · Saumya Sinha · Qinghu Tang · Armi Tiihonen · Ruben Glatt · Muge Komurcu · Jan Drgona · Juan Gomez-Romero · Ashish Kapoor · Dylan J Fitzpatrick · Alireza Rezvanifar · Adrian Albert · Olya (Olga) Irzak · Kara Lamb · Ankur Mahesh · Kiwan Maeng · Frederik Kratzert · Sorelle Friedler · Niccolo Dalmasso · Alex Robson · Lindiwe Malobola · Lucas Maystre · Yu-wen Lin · Surya Karthik Mukkavili · Brian Hutchinson · Alexandre Lacoste · Yanbing Wang · Zhengcheng Wang · Yinda Zhang · Victoria Preston · Jacob Pettit · Draguna Vrabie · Miguel Molina-Solana · Tonio Buonassisi · Andrew Annex · Tunai P Marques · Catalin Voss · Johannes Rausch · Max Evans