Non-Asymptotic Analysis of Robust Control from Coarse-Grained Identification

Abstract

This work explores the trade-off between the number of samples required to accurately build models of dynamical systems and the degradation of performance in various control objectives due to a coarse approximation. In particular, we show that simple models can be easily fit from input/output data and are sufficient for achieving various control objectives. We derive bounds on the number of noisy input/output samples from a stable linear time-invariant system that are sufficient to guarantee that the corresponding finite impulse response approximation is close to the true system in the ${\mathcal{H}}_{\infty }$-norm. We demonstrate that these demands are lower than those derived in prior art which aimed to accurately identify dynamical models. We also explore how different physical input constraints, such as power constraints, affect the sample complexity. Finally, we show how our analysis fits within the established framework of robust control, by demonstrating how a controller designed for an approximate system provably meets performance objectives on the true system.

arXiv commentA substantial revision, where we strengthen our existing upper bounds and introduce a matching lower bound

References in corpus (3)

• Gradient Descent Learns Linear Dynamical Systems
• Atomic norm denoising with applications to line spectral estimation
• Exponential Stability Analysis via Integral Quadratic Constraints

Cited by in corpus (21)

• On the Sample Complexity of the Linear Quadratic Regulator
• Stable Recurrent Models
• Non-asymptotic Identification of Linear Dynamical Systems Using Multiple Trajectories
• numpywren: serverless linear algebra
• Learning Without Mixing: Towards A Sharp Analysis of Linear System Identification
• Certainty Equivalence is Efficient for Linear Quadratic Control
• On the sample complexity of stabilizing linear dynamical systems from data
• Sample Complexity of Sparse System Identification Problem
• No-Regret Prediction in Marginally Stable Systems
• Episodic Linear Quadratic Regulators with Low-rank Transitions
• Nonparametric Finite Time LTI System Identification
• Robust guarantees for learning an autoregressive filter
• Towards a Dimension-Free Understanding of Adaptive Linear Control
• Task-Optimal Exploration in Linear Dynamical Systems
• Learning Partially Observed Linear Dynamical Systems from Logarithmic Number of Samples
• Minimax Lower Bounds for ${\mathcal{H}}_{\infty }$-Norm Estimation
• Sample Complexity of the Robust LQG Regulator with Coprime Factors Uncertainty
• Certainty Equivalent Quadratic Control for Markov Jump Systems
• Data Driven Estimation of Stochastic Switched Linear Systems of Unknown Order
• Improved rates for prediction and identification of partially observed linear dynamical systems
• Identification and Adaptive Control of Markov Jump Systems: Sample Complexity and Regret Bounds