Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Resilient control under denial-of-service and uncertainty: An adaptive dynamic programming approach

Weinan Gao, Zhong-Ping Jiang, Tianyou Chai

TL;DR

Using techniques from reinforcement learning and output regulation theory, it is shown that resilient optimal controllers can be learned directly from real-time state and input data collected from the systems subjected to attacks.

Abstract

In this paper, a new framework for the resilient control of continuous-time linear systems under denial-of-service (DoS) attacks and system uncertainty is presented. Integrating techniques from reinforcement learning and output regulation theory, it is shown that resilient optimal controllers can be learned directly from real-time state and input data collected from the systems subjected to attacks. Sufficient conditions are given under which the closed-loop system remains stable given any upper bound of DoS attack duration. Simulation results are used to demonstrate the efficacy of the proposed learning-based framework for resilient control under DoS attacks and model uncertainty.

Resilient control under denial-of-service and uncertainty: An adaptive dynamic programming approach

TL;DR

Using techniques from reinforcement learning and output regulation theory, it is shown that resilient optimal controllers can be learned directly from real-time state and input data collected from the systems subjected to attacks.

Abstract

In this paper, a new framework for the resilient control of continuous-time linear systems under denial-of-service (DoS) attacks and system uncertainty is presented. Integrating techniques from reinforcement learning and output regulation theory, it is shown that resilient optimal controllers can be learned directly from real-time state and input data collected from the systems subjected to attacks. Sufficient conditions are given under which the closed-loop system remains stable given any upper bound of DoS attack duration. Simulation results are used to demonstrate the efficacy of the proposed learning-based framework for resilient control under DoS attacks and model uncertainty.

Paper Structure

This paper contains 8 sections, 3 theorems, 32 equations, 3 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. Problem Formulation and Preliminaries
  3. Linear Optimal Output Regulation
  4. DoS Attack Model and Resilient Optimal Control
  5. Main Results
  6. Learning-based Controller Design Under DoS
  7. Example
  8. Conclusion

Key Result

Lemma 1

For $i=0,1,2,\cdots,(n-r)q+1$, if there exists a $s^*\in\mathbb{Z}_+$ such that for all $s>s^*$, then, the following properties hold

Figures (3)

  • Figure 1: Learning-based resilient optimal control framework
  • Figure 2: Comparison between the optimal control gain $K^*$ and the learned control gain $K_k$ at each iteration
  • Figure 3: Simulation results: (a) Acceleration profile of preceding vehicle, (b) Clearance error of autonomous vehicle using proposed control method, internal-model-based control method, and model-based resilient control method.

Theorems & Definitions (9)

  • Remark 1
  • Definition 1
  • Lemma 1
  • Lemma 2
  • Remark 2
  • Remark 3
  • Theorem 1
  • Remark 4
  • Remark 5