Corpus ID: 209405012

Learning the Globally Optimal Distributed LQ Regulator

  title={Learning the Globally Optimal Distributed LQ Regulator},
  author={Luca Furieri and Y. Zheng and M. Kamgarpour},
  • Luca Furieri, Y. Zheng, M. Kamgarpour
  • Published 2019
  • Computer Science, Engineering, Mathematics
  • ArXiv
  • We study model-free learning methods for the output-feedback Linear Quadratic (LQ) control problem in finite-horizon subject to subspace constraints on the control policy. Subspace constraints naturally arise in the field of distributed control and present a significant challenge in the sense that standard model-based optimization and learning leads to intractable numerical programs in general. Building upon recent results in zeroth-order optimization, we establish model-free sample-complexity… CONTINUE READING
    Distributed Online Linear Quadratic Control for Linear Time-invariant Systems
    Learning Partially Observed Linear Dynamical Systems from Logarithmic Number of Samples


    Publications referenced by this paper.
    A Characterization of Convex Problems in Decentralized Control$^ast$
    • 640
    • Highly Influential
    • PDF
    On the Sample Complexity of the Linear Quadratic Regulator
    • 198
    • Highly Influential
    • PDF
    Regret Bounds for the Adaptive Control of Linear Quadratic Systems
    • 160
    • Highly Influential
    • PDF
    Global Convergence of Policy Gradient Methods for the Linear Quadratic Regulator
    • 145
    • Highly Influential
    • PDF
    Regret Bounds for Robust Adaptive Control of the Linear Quadratic Regulator
    • 99
    • PDF
    Stochastic First- and Zeroth-Order Methods for Nonconvex Stochastic Programming
    • 574
    • PDF
    Modern Wiener-Hopf Design of Optimal Controllers. Part I
    • 1,371
    • PDF
    Convex Relaxation for Optimal Distributed Control Problems
    • 80
    • PDF
    Derivative-Free Methods for Policy Optimization: Guarantees for Linear Quadratic Systems
    • 58
    • Highly Influential
    • PDF
    Model-Free Linear Quadratic Control via Reduction to Expert Prediction
    • 53
    • PDF