Separation of learning and control for cyber-physical systems
From MaRDI portal
Abstract: Most cyber-physical systems (CPS) encounter a large volume of data which is added to the system gradually in real time and not altogether in advance. In this paper, we provide a theoretical framework that yields optimal control strategies for such CPS at the intersection of control theory and learning. In the proposed framework, we use the actual CPS, i.e., the "true" system that we seek to optimally control online, in parallel with a model of the CPS that is available. We then institute an information state for the system which does not depend on the control strategy. An important consequence of this independence is that for any given choice of a control strategy and a realization of the system's variables until time t, the information states at future times do not depend on the choice of the control strategy at time t but only on the realization of the decision at time t, and thus they are related to the concept of separation between estimation of the state and control. Namely, the future information states are separated from the choice of the current control strategy. Such control strategies are called separated control strategies. Hence, we can derive offline the optimal control strategy of the system with respect to the information state, which might not be precisely known due to model uncertainties or complexity of the system, and then use standard learning approaches to learn the information state online while data are added gradually to the system in real time. We show that after the information state becomes known, the separated control strategy of the CPS model derived offline is optimal for the actual system. We illustrate the proposed framework in a dynamic system consisting of two subsystems with a delayed sharing information structure.
Recommendations
- Optimal control and learning for <scp>cyber‐physical</scp> systems
- Networked learning predictive control of nonlinear cyber-physical systems
- scientific article; zbMATH DE number 1254130
- scientific article; zbMATH DE number 21854
- A real-time big data control-theoretical framework for cyber-physical-human systems
Cites work
- A data‐based private learning framework for enhanced security against replay attacks in cyber‐physical systems
- A Duality Framework for Stochastic Optimal Control of Complex Systems
- A General Safety Framework for Learning-Based Control in Uncertain Robotic Systems
- A separation theorem for periodic sharing information patterns in decentralized control
- A standard form for sequential stochastic control
- An example of interaction between information and control: The Transparency of a game
- Coordination control of distributed systems
- Corrections and extensions to "Decentralized stochastic control with delayed sharing information pattern"
- Decentralized optimal control of Markov chains with a common past information set
- Decentralized Q-Learning for Stochastic Teams and Games
- Decentralized Stochastic Control with Partial History Sharing: A Common Information Approach
- Dynamic programming and optimal control. Vol. 1.
- scientific article; zbMATH DE number 3148886 (Why is no real title available?)
- scientific article; zbMATH DE number 4160608 (Why is no real title available?)
- scientific article; zbMATH DE number 5562427 (Why is no real title available?)
- scientific article; zbMATH DE number 47864 (Why is no real title available?)
- scientific article; zbMATH DE number 47865 (Why is no real title available?)
- scientific article; zbMATH DE number 3460142 (Why is no real title available?)
- scientific article; zbMATH DE number 1321699 (Why is no real title available?)
- scientific article; zbMATH DE number 2107836 (Why is no real title available?)
- scientific article; zbMATH DE number 810021 (Why is no real title available?)
- scientific article; zbMATH DE number 7625165 (Why is no real title available?)
- scientific article; zbMATH DE number 3376944 (Why is no real title available?)
- Iterative learning control based on extremum seeking
- Learning Model Predictive Control for Iterative Tasks. A Data-Driven Control Framework
- Nonparametric Estimation of Conditional Distributions
- On delayed sharing patterns
- On Social Optimal Routing Under Selfish Learning
- On Team Decision Problems with Nonclassical Information Structures
- On the complexity of decentralized decision making and detection problems
- On the complexity of designing distributed protocols
- On the existence of optimal policies for a class of static and sequential dynamic teams
- Online learning of Nash equilibria in congestion games
- Optimal Control Strategies in Delayed Sharing Information Structures
- Partially observed Markov decision processes. From filtering to controlled sensing
- Provably safe and robust learning-based model predictive control
- Statistical learning for analysis of networked control systems over unknown channels
- Sufficient statistics in the optimum control of stochastic systems
- The Complexity of Markov Decision Processes
Cited in
(5)- A real-time big data control-theoretical framework for cyber-physical-human systems
- Adaptive control for cyber-physical systems under man-in-the-middle attacks with false data injections
- Combining learning and control in linear systems
- Stochastic control with distributionally robust constraints for cyber-physical systems vulnerable to attacks
- Optimal tampering attack strategy for FIR system identification with multi-level quantized observations
This page was built for publication: Separation of learning and control for cyber-physical systems
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6164023)
