Identification and adaptation with binary-valued observations under non-persistent excitation condition

DOI10.1016/J.AUTOMATICA.2022.110158zbMATH Open1485.93607arXiv2107.03588OpenAlexW3181955781MaRDI QIDQ2123231FDOQ2123231

Authors: Yanyan Li 

Publication date: 8 April 2022

Published in: Automatica (Search for Journal in Brave)

Abstract: Dynamical systems with binary-valued observations are widely used in information industry, technology of biological pharmacy and other fields. Though there have been much efforts devoted to the identification of such systems, most of the previous investigations are based on first-order gradient algorithm which usually has much slower convergence rate than the Quasi-Newton algorithm. Moreover, persistence of excitation(PE) conditions are usually required to guarantee consistent parameter estimates in the existing literature, which are hard to be verified or guaranteed for feedback control systems. In this paper, we propose an online projected Quasi-Newton type algorithm for parameter estimation of stochastic regression models with binary-valued observations and varying thresholds. By using both the stochastic Lyapunov function and martingale estimation methods, we establish the strong consistency of the estimation algorithm and provide the convergence rate, under a signal condition which is considerably weaker than the traditional PE condition and coincides with the weakest possible excitation known for the classical least square algorithm of stochastic regression models. Convergence of adaptive predictors and their applications in adaptive control are also discussed.

Full work available at URL: https://arxiv.org/abs/2107.03588

Recommendations

• Recursive projection algorithm on FIR system identification with binary-valued observations
• A unified identification algorithm of FIR systems based on binary observations with time-varying thresholds
• Parameter estimation in systems with binary-valued observations and structural uncertainties
• Convergence analysis of an online approach to parameter estimation problems based on binary observations
• Distributed recursive projection identification with binary-valued observations

zbMATH Keywords

identificationadaptationmartingalesbinary-valued observationquasi-Newton algorithmpersistent excitation

Mathematics Subject Classification ID

Estimation and detection in stochastic control theory (93E10) Identification in stochastic control theory (93E12) Stochastic learning and adaptive control (93E35)

Cites Work

• Least squares estimates in stochastic regression models with applications to identification and control of dynamic systems
• Identification of ARMA models using intermittent and quantized output observations
• Recursive algorithms for parameter estimation with adaptive quantizer
• Recursive projection algorithm on FIR system identification with binary-valued observations
• Identification of Wiener systems with binary-valued output observations
• Asymptotically efficient parameter estimation using quantized output observations
• Joint identification of plant rational models and noise distribution functions using binary-valued observations
• Title not available (Why is that?)
• Title not available (Why is that?)
• Identification and stochastic adaptive control
• System identification using binary sensors
• Adaptive Tracking Control of A Class of First-Order Systems With Binary-Valued Observations and Time-Varying Thresholds
• Convergence and logarithm laws of self-tuning regulators
• Asymptotic properties of least-squares estimates in stochastic regression models
• Asymptotically efficient adaptive control in stochastic regression models
• Identification of FIR Systems Based on Quantized Output Measurements: A Quadratic Programming-Based Method
• On strong consistency of least squares identification algorithms
• A new kernel-based approach to system identification with quantized output data
• Recursive identification of systems with binary-valued outputs and with ARMA noises
• Adaptive Tracking Control of Linear Systems With Binary-Valued Observations and Periodic Target

Cited In (13)

• A unified identification algorithm of FIR systems based on binary observations with time-varying thresholds
• Asymptotically efficient quasi-Newton type identification with quantized observations under bounded persistent excitations
• System identification: regime switching, unmodeled dynamics, and binary sensors
• Parameter estimation in systems with binary-valued observations and structural uncertainties
• Estimation of IIR systems with binary-valued observations
• Recursive projection algorithm on FIR system identification with binary-valued observations
• Adaptive tracking of a class of first-order systems with binary-valued observations and fixed thresholds
• Joint identification of system parameter and noise parameters in quantized systems
• Distributed recursive projection identification with binary-valued observations
• Binary observation-based FIR system identification against replay attacks
• Optimal adaptive identification under saturated output observations and non-i.i.d data
• Identification of FIR systems with binary-valued observations against denial-of-service attacks
• Threshold selection and resource allocation for quantized identification

This page was built for publication: Identification and adaptation with binary-valued observations under non-persistent excitation condition

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2123231)