Adaptive Control of Unknown Linear Switched Systems via Policy Gradient Methods
On the map: Deep reinforcement learning methods → RL for games and combinatorial optimization
Abstract (arXiv)
We consider the policy gradient adaptive control (PGAC) framework, which adaptively updates a control policy in real time, by performing data-based gradient descent steps on the linear quadratic regulator cost. This method has empirically shown to react to changing circumstances, such as model parameters, efficiently. To formalize this observation, we design a PGAC method which stabilizes linear switched systems, where both model parameters and switching time are unknown. We use sliding window data for the policy gradient estimate and show that under a dwell time condition and small dynamics variation, the policy can track the switching dynamics and ensure closed-loop stability. We perform simulations to validate our theoretical results.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in RL for games and combinatorial optimization, one of 15 directions in Deep reinforcement learning methods. Builds on 19 of its 31 references that are in the map. First posted on arXiv 6 Jan 2026.
Builds on
19 of 31 references.
Closest ancestors · 1
Same direction, newest first
Other references · 18
- 2018Global Convergence of Policy Gradient Methods for the Linear Quadratic Regulatorcited by 250
- 2020Naive Exploration is Optimal for Online LQRcited by 116
- 2019Learning Linear-Quadratic Regulators Efficiently with only √T Regretcited by 70
- 2018Online Linear Quadratic Controlcited by 56
- 2022Statistical Learning Theory for Control: A Finite Sample Perspectivecited by 53
- 2019Formulas for Data-driven Control: Stabilization, Optimality and Robustnesscited by 47
- 2019Convergence and sample complexity of gradient methods for the model-free linear quadratic regulator problemcited by 46
- 2018Data-Enabled Predictive Control: In the Shallows of the DeePCcited by 38
- 2019Data-Driven Model Predictive Control with Stability and Robustness Guaranteescited by 26
- 2019Data informativity: a new perspective on data-driven analysis and controlcited by 19
- 2024Data-Enabled Policy Optimization for Direct Adaptive Learning of the LQRcited by 9
- 2021On the Certainty-Equivalence Approach to Direct Data-Driven LQR Designcited by 5
6 more
- 2020Exact Asymptotics for Linear Quadratic Adaptive Controlcited by 4
- 2021Online learning of data-driven controllers for unknown switched linear systemscited by 3
- 2024Direct Adaptive Control of Grid-Connected Power Converters via Output-Feedback Data-Enabled Policy Optimizationcited by 2
- 2022Data-driven stabilization of switched and constrained linear systemscited by 1
- 2023An Online Adaptation Strategy for Direct Data-driven Controlcited by 1
- 2024Online Data-Driven Adaptive Control for Unknown Linear Time-Varying Systemscited by 1
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2026Adaptive Linear Quadratic Control of Unknown Linear Time-Varying Systems via Policy Gradient Methods ↗
- 2026Model-Free Output Feedback Stabilization via Policy Gradient Methods
- 2025Perturbed Gradient Descent Algorithms are Small-Disturbance Input-to-State Stable
- 2020Learning Constrained Adaptive Differentiable Predictive Control Policies With Guarantees
- 2024Stability-Certified On-Policy Data-Driven LQR via Recursive Learning and Policy Gradient
- 2026A Stochastic Gradient Descent Approach to Design Policy Gradient Methods for LQR
- 2026Gradient Dominance in the Linear Quadratic Regulator: A Unified Analysis for Continuous-Time and Discrete-Time Systems
- 2024Policy Gradient-based Model Free Optimal LQG Control with a Probabilistic Risk Constraint