Corpus ID: 235457972

Robustness and Consistency in Linear Quadratic Control with Predictions

  title={Robustness and Consistency in Linear Quadratic Control with Predictions},
  author={Tongxin Li and Ruixiao Yang and Guannan Qu and Guanya Shi and Chenkai Yu and Adam Wierman and Steven H. Low},
We study the problem of learning-augmented predictive linear quadratic control. Our goal is to design a controller that balances consistency, which measures the competitive ratio when predictions are accurate, and robustness, which bounds the competitive ratio when predictions are inaccurate. We propose a novel λconfident controller and prove that it maintains a competitive ratio upper bound of 1 +min{O(λ2ε) +O(1− λ)2, O(1) +O(λ2)} where λ ∈ [0, 1] is a trust parameter set based on the… Expand

Figures from this paper


Regret Bounds for the Adaptive Control of Linear Quadratic Systems
The construction of the condence set is based on the recent results from online least-squares estimation and leads to improved worst-case regret bound for the proposed algorithm, and is the the rst time that a regret bound is derived for the LQ control problem. Expand
On the Sample Complexity of the Linear Quadratic Regulator
This paper proposes a multi-stage procedure that estimates a model from a few experimental trials, estimates the error in that model with respect to the truth, and then designs a controller using both the model and uncertainty estimate, and provides end-to-end bounds on the relative error in control cost. Expand
Online Optimal Control with Linear Dynamics and Predictions: Algorithms and Regret Analysis
This paper designs online algorithms, Receding Horizon Gradient-based Control (RHGC), that utilize the predictions through finite steps of gradient computations, and provides a fundamental limit of the dynamic regret for any online algorithms by considering linear quadratic tracking problems. Expand
Optimal Robustness-Consistency Trade-offs for Learning-Augmented Online Algorithms
This paper provides the first set of non-trivial lower bounds for competitive analysis using machine-learned predictions on the classic problems of ski-rental and non-clairvoyant scheduling and provides optimal trade-offs in various settings. Expand
Online Control with Adversarial Disturbances
The objective is to desire an online control procedure that can do nearly as well as that of a procedure that has full knowledge of the disturbances in hindsight, and the main result is an efficient algorithm that provides nearly tight regret bounds for this problem. Expand
The Primal-Dual method for Learning Augmented Algorithms
This paper extends the primal-dual method for online algorithms in order to incorporate predictions that advise the online algorithm about the next action to take and uses this framework to obtain novel algorithms for a variety of online covering problems. Expand
Follow-the-Regularized-Leader and Mirror Descent: Equivalence Theorems and L1 Regularization
It is proved that many mirror descent algorithms for online convex optimization (such as online gradient descent) have an equivalent interpretation as follow-the-regularizedleader (FTRL) algorithms, and the FTRL-Proximal algorithm can be seen as a hybrid of these two algorithms, which significantly outperforms both on a large, realworld dataset. Expand
A Course in Robust Control Theory: A Convex Approach
0 Introduction.- 1 Preliminaries in Finite Dimensional Space.- 2 State Space System Theory.- 3 Linear Analysis.- 4 Model Realizations and Reduction.- 5 Stabilizing Controllers.- 6 H2 OptimalExpand
Robust model predictive control: A survey
The basic concepts of MPC are reviewed, the uncertainty descriptions considered in the MPC literature are surveyed, and the techniques proposed for robust constraint handling, stability, and performance are surveyed. Expand
Competitive caching with machine learned advice
This work develops a framework for augmenting online algorithms with a machine learned oracle to achieve competitive ratios that provably improve upon unconditional worst case lower bounds when the oracle has low error. Expand