On the Equivalence of OKID and Time Series Identification for Markov-Parameter Estimation

Size: px
Start display at page:

Download "On the Equivalence of OKID and Time Series Identification for Markov-Parameter Estimation"

Transcription

1 On the Equivalence of OKID and Time Series Identification for Markov-Parameter Estimation P V Albuquerque, M Holzel, and D S Bernstein April 5, 2009 Abstract We show the equivalence of Observer/Kalman Identification OKID and Time Series Identification TSI for Markov parameters PhD Student, University of Michigan, Ann Arbor, MI PhD Student, University of Michigan, Ann Arbor, MI Professor, Aerospace Engineering Department, University of Michigan, Ann Arbor, MI, 48109

2 1 Introduction Introduce paper here 2 Markov Parameter Estimation using OKID Observer/Kalman System Identification OKID involves adding an observer into an observable discrete-time system in order to help determine the Markov parameters The method for doing so, as seen in [2, is clearly derived in this section 21 Discrete-Time Linear System Consider the linear discrete-time system xk + 1 = Axk + Buk, 1 yk = Cxk + Duk, 2 k 0, xk R n, yk R p, and uk R m, and A, B, C, and D are real matrices of corresponding sizes We assume that A, C is observable For nonnegative integers k and r we have xk + r = A k xr + yk + r = CA k xr + A i 1 Buk + r i, 3 CA i 1 Buk + r i + Duk + r, 4 0 = 0 Now let s r Then, lining up 4 for k = 0,, s r yields Y [r:s = G [0:s r X r,s + H [0:s r U r,s, 5 Y [r:s [ yr yr + 1 yr + 2 ys R p s r+1, G [0:s r [ C CA CA 2 CA s r R p ns r+1, H [0:s r [ H 0 H 1 H 2 H s r [ D CB CAB CA s r 1 B R p ms r+1, 2

3 xr xr X r,s 0 Rns r+1 s r+1, 0 0 xr ur ur + 1 us 0 ur us 1 U r,s Rms r+1 s r ur 22 Adding an Observer to the System Adding and subtracting F yk, F R n p, to the right hand side of 1 yields which, with 2, can be written as xk + 1 = Axk + Buk + F yk F yk 6 = A + F Cxk + B + F Duk F yk, 7 xk + 1 = Axk + Bvk, 8 yk = Cxk + Dvk, 9 A A + F C R n n, B [B + F D F R n m+p, D [D 0 R p m+p, [ uk vk R m+p yk For nonnegative integers k and r we have xk + r = A k xr + yk + r = CA k xr + A i 1 Bvk + r i, 10 CA i 1 Bvk + r i + Dvk + r 11 Now, let s r Then, in analogy with 5, we have Y [r:s = G [0:s r X r,s + H [0:s r V r,s, 12 3

4 G [0,s r [ C CA CA 2 CA s r R p ns r+1, H [0:s r [ H 0 H 1 H 2 H s r [ D CB CAB CA s r 1 B R p m+ps r+1, vr vr + 1 vs 0 vr vs 1 V r,s Rm+ps r+1 s r vr Let r 0 and r + n < s so that n < s r Then, Y [r:s = [ Y [r:r+n 1 Y [r+n:s R p s r+1, G [0:s r = [ G [0:n 1 G [n:s r R p ns r+1, H [0:s r = [ H [0:n+1 H [n:s r R p m+ps r+1, V r,s = vr vr + 1 vr + n 1 vr + n vr + n + 1 vs 1 vs 0 vr vr + n 2 vr + n 1 vr + n vs 2 vs vr vr + 1 vr + 2 vs n vs n vr vr + 1 vs n 1 vs n vr vs n vs n vr vs n vr V r,r+n 1 V r+n,s 0 m+p n 0 m+ps n r n 0 m+ps n r 1 V r,s n 1 R m+ps r+1 s r+1, 4

5 V r+n,s R m+pn+1 s r n+1 is given by vr + n vr + 2n V r+n,s v2r + 2n s vr + n, s < r + 2n, vr vs n vr + n vr + 2n V r+n,s, s = r + 2n, vr vr + n vr + n vr + 2n vs V r+n,s, s > r + 2n, vr vr + n vs n Note that for s = n + 1, and thus r = 0, V 0,0 denotes v0 From 12, we have Y [r+n:s = G [n:s r X r+n,s + H [0:n+1 V r+n,s + H [n:s r [0 m+ps n r 1 V r,s n 1, 13 Note that, since H 0 = D = [D 0, D can be replaced by D, and thus columns p + 1,, 2p of H [0:n+1 and rows m + 1,, m + p of V r+n,s can be deleted In addition, by defining H [0:n+1 [ D H 1 H 2 H n+1 R p m+pn+m, un + r un r us V vn 1 + r vn + r vs 1 r+n,s Rm+pn+m s r n+1 vr v1 + r vs n Since A, C is observable, we choose F such that A is nilpotent Hence for all k n, it follows that A k = 0 so that H k = 0 for all k n + 2 Therefore, for k n, 10 and 11 can be written as xk + r = yk + r = minn,k minn,k A i 1 Bvk + r i, 14 CA i 1 Bvk + r i + Dvk + r 15 5

6 Based on the nilpotent assumption, G [n,s r = 0 p ns r n+1, H [n:s r = 0 p m+ps r n 1 Thus 13 can be rewritten as Y [r+n:s = H [0:n+1V r+n,s 16 If V r+n,s has full column rank, then the unique solution of 16 is given by H [0:n+1 = Y [r+n:s V + r+n,s, 17 + is the generalized inverse When u and y are corrupted by noise, 17 provides a least squares estimate 23 Deriving H i from H [0:n+1 Let k 1 Then H k = CA k 1 B = [β k α k, β k CA + F C k 1 B + F D, 18 α k CA + F C k 1 F 19 Next, for k 3 we have H k = CA k 1 B, 20 = C A + F C F C A k 2 B = C A + F CA F CA A k 3 B = C A + F CA + F C F C F CA A k 3 B = C A + F C 2 A + F CF C F CA A k 3 B 2 = C A + F C 2 A + F C i 1 F CA 2 i A k 3 B 21 6

7 More generally we have H k = C A + F C j j A + F C i 1 F CA j i A k j+1 B = C A + F C j A j A + F C i 1 F CA j+1 i A k j+2 B = C A + F C j A + F C F C j A + F C i 1 F CA j+1 i A k j+2 B = C A + F C j+1 A + F C j F C j A + F C i 1 F CA j+1 i A k j+2 B = C A + F C j+1 j+1 A + F C i 1 F CA j i A k j+2 B 22 Using 22, 21 becomes k 1 H k = C A + F C k 1 A + F C i 1 F CA k i 1 B k 1 = CA + F C k 1 B C A + F C i 1 F CA k i 1 B = α k D + CA + F C k 1 B k 1 C A + F C i 1 F CA k i 1 B + α k D = β k k 1 CA + F C i 1 F CA k i 1 B + α k D k 1 = β k α i H k i + α k H 0 Thus, β k + k α i H k i, k = 1,, n; H k = n α i H k i, k n

8 3 Alternative Derivation of OKID Using a Pseudo- FIR Model The discrete-time state space model 1-2 can be represented by the infinite impulse response IIR model y k = α 1 y k α n y k n + β 0 u k + β 1 u k β n u k n, 24 for k n Rearranging 24, we obtain y k = H 0 v k + + H n v k n, 25 H k [ β k α k, 26 and from 24, α 0 = 0 Furthermore, note that 25 can be written as which has the least-squares solution and ˆα 0 is constrained to be zero Y [r:s = H [0:s r V r:s, 27 Ĥ [0:s r = V + r:sy [r:s, 28 Ĥ k [ ˆβk ˆα k, 29 Next, note that if y 0,, y µ is the impulse response of a system, then y k H k Impulsing 24, we obtain H 0 = β 0, H 1 = α 1 y 0 + β 1 u 0 = α 1 H 0 + β 1, H 2 = α 1 y 1 + α 2 y 0 + β 2 u 0 = α 1 H 1 + α 2 H 0 + β 2 H 3 = α 1 y 2 + α 2 y 1 + α 3 y 0 + β 3 u 0 = α 1 H 2 + α 2 H 1 + α 3 H 0 + β 3 β k + k α k H k i, 0 k n, H k = α k H k i, k > n 8

9 Thus using the estimates 29, the Markov parameters can be estimated recursively by ˆβ k + k ˆα k Ĥ k i, 0 k n, Ĥ k = 30 ˆα k Ĥ k i, k > n 4 Conclusion Note that 17 and 28 are reshuffled representations of the same equation While OKID and TSI both set up the V matrix differently, each contains the same information and thus the least square approximation of H must be identical This conclusion is verified in [1 Also, OKID and TSI share a very similar method of obtaining the estimates of the Markov parameters in 23 and 30 References [1 S Barnett Inversion of partitioned matrices with patterned blocks International Journal of Systems Science, 142: , 1983 [2 ; Wei Chen John Valasek Observer/kalman filter identification for online system identification of aircraft Journal of Guidance, Control, and Dynamics, : ,

An All-Interaction Matrix Approach to Linear and Bilinear System Identification

An All-Interaction Matrix Approach to Linear and Bilinear System Identification An All-Interaction Matrix Approach to Linear and Bilinear System Identification Minh Q. Phan, Francesco Vicario, Richard W. Longman, and Raimondo Betti Abstract This paper is a brief introduction to the

More information

Observers for Bilinear State-Space Models by Interaction Matrices

Observers for Bilinear State-Space Models by Interaction Matrices Observers for Bilinear State-Space Models by Interaction Matrices Minh Q. Phan, Francesco Vicario, Richard W. Longman, and Raimondo Betti Abstract This paper formulates a bilinear observer for a bilinear

More information

Abstract. 1 Introduction

Abstract. 1 Introduction Relationship between state-spaceand input-output models via observer Markov parameters M.Q. Phan,' R. W. Longman* ^Department of Mechanical and Aerospace Engineering, Princeton University, Princeton, NJ

More information

Generalized Framework of OKID for Linear State-Space Model Identification

Generalized Framework of OKID for Linear State-Space Model Identification Generalized Framework of OKID for Linear State-Space Model Identification Francesco Vicario, Minh Q. Phan, Richard W. Longman, and Raimondo Betti Abstract This paper presents a generalization of observer/kalman

More information

Observers for Linear Systems with Unknown Inputs

Observers for Linear Systems with Unknown Inputs Chapter 3 Observers for Linear Systems with Unknown Inputs As discussed in the previous chapters, it is often the case that a dynamic system can be modeled as having unknown inputs (e.g., representing

More information

Advanced Process Control Tutorial Problem Set 2 Development of Control Relevant Models through System Identification

Advanced Process Control Tutorial Problem Set 2 Development of Control Relevant Models through System Identification Advanced Process Control Tutorial Problem Set 2 Development of Control Relevant Models through System Identification 1. Consider the time series x(k) = β 1 + β 2 k + w(k) where β 1 and β 2 are known constants

More information

Modelling and Control of Dynamic Systems. Stability of Linear Systems. Sven Laur University of Tartu

Modelling and Control of Dynamic Systems. Stability of Linear Systems. Sven Laur University of Tartu Modelling and Control of Dynamic Systems Stability of Linear Systems Sven Laur University of Tartu Motivating Example Naive open-loop control r[k] Controller Ĉ[z] u[k] ε 1 [k] System Ĝ[z] y[k] ε 2 [k]

More information

System Identification Using a Retrospective Correction Filter for Adaptive Feedback Model Updating

System Identification Using a Retrospective Correction Filter for Adaptive Feedback Model Updating 9 American Control Conference Hyatt Regency Riverfront, St Louis, MO, USA June 1-1, 9 FrA13 System Identification Using a Retrospective Correction Filter for Adaptive Feedback Model Updating M A Santillo

More information

Position Control Using Acceleration- Based Identification and Feedback With Unknown Measurement Bias

Position Control Using Acceleration- Based Identification and Feedback With Unknown Measurement Bias Position Control Using Acceleration- Based Identification and Feedback With Unknown Measurement Bias Jaganath Chandrasekar e-mail: jchandra@umich.edu Dennis S. Bernstein e-mail: dsbaero@umich.edu Department

More information

16.30 Estimation and Control of Aerospace Systems

16.30 Estimation and Control of Aerospace Systems 16.30 Estimation and Control of Aerospace Systems Topic 5 addendum: Signals and Systems Aeronautics and Astronautics Massachusetts Institute of Technology Fall 2010 (MIT) Topic 5 addendum: Signals, Systems

More information

BATCH-FORM SOLUTIONS TO OPTIMAL INPUT SIGNAL RECOVERY IN THE PRESENCE OF NOISES

BATCH-FORM SOLUTIONS TO OPTIMAL INPUT SIGNAL RECOVERY IN THE PRESENCE OF NOISES (Preprint) AAS BATCH-FORM SOLUTIONS TO OPTIMAL INPUT SIGNAL RECOVERY IN THE PRESENCE OF NOISES P Lin, M Q Phan, and S A Ketcham This paper studies the problem of optimally recovering the input signals

More information

Discrete Time Systems:

Discrete Time Systems: Discrete Time Systems: Finding the control sequence {u(k): k =, 1, n} that will drive a discrete time system from some initial condition x() to some terminal condition x(k). Given the discrete time system

More information

Cumulative Retrospective Cost Adaptive Control with RLS-Based Optimization

Cumulative Retrospective Cost Adaptive Control with RLS-Based Optimization 21 American Control Conference Marriott Waterfront, Baltimore, MD, USA June 3-July 2, 21 ThC3.3 Cumulative Retrospective Cost Adaptive Control with RLS-Based Optimization Jesse B. Hoagg 1 and Dennis S.

More information

4F3 - Predictive Control

4F3 - Predictive Control 4F3 Predictive Control - Lecture 3 p 1/21 4F3 - Predictive Control Lecture 3 - Predictive Control with Constraints Jan Maciejowski jmm@engcamacuk 4F3 Predictive Control - Lecture 3 p 2/21 Constraints on

More information

Kalman-Filter-Based Time-Varying Parameter Estimation via Retrospective Optimization of the Process Noise Covariance

Kalman-Filter-Based Time-Varying Parameter Estimation via Retrospective Optimization of the Process Noise Covariance 2016 American Control Conference (ACC) Boston Marriott Copley Place July 6-8, 2016. Boston, MA, USA Kalman-Filter-Based Time-Varying Parameter Estimation via Retrospective Optimization of the Process Noise

More information

CDS Final Exam

CDS Final Exam CDS 22 - Final Exam Instructor: Danielle C. Tarraf December 4, 2007 INSTRUCTIONS : Please read carefully! () Description & duration of the exam: The exam consists of 6 problems. You have a total of 24

More information

Lecture 2 Discrete-Time LTI Systems: Introduction

Lecture 2 Discrete-Time LTI Systems: Introduction Lecture 2 Discrete-Time LTI Systems: Introduction Outline 2.1 Classification of Systems.............................. 1 2.1.1 Memoryless................................. 1 2.1.2 Causal....................................

More information

IDENTIFICATION AND CONTROL OF A DISTILLATION COLUMN. Department of Informatics and Systems University of Pavia, ITALY

IDENTIFICATION AND CONTROL OF A DISTILLATION COLUMN. Department of Informatics and Systems University of Pavia, ITALY IDENTIFICATION AND CONTROL OF A DISTILLATION COLUMN Antonio Tiano (1), Antonio Zirilli (2) (1) Department of Informatics and Systems University of Pavia, ITALY Email: antonio@control1.unipv.it (2) Honeywell

More information

Rural/Urban Migration: The Dynamics of Eigenvectors

Rural/Urban Migration: The Dynamics of Eigenvectors * Analysis of the Dynamic Structure of a System * Rural/Urban Migration: The Dynamics of Eigenvectors EGR 326 April 11, 2019 1. Develop the system model and create the Matlab/Simulink model 2. Plot and

More information

Closed-Loop Identification of Unstable Systems Using Noncausal FIR Models

Closed-Loop Identification of Unstable Systems Using Noncausal FIR Models 23 American Control Conference (ACC) Washington, DC, USA, June 7-9, 23 Closed-Loop Identification of Unstable Systems Using Noncausal FIR Models Khaled Aljanaideh, Benjamin J. Coffer, and Dennis S. Bernstein

More information

1. Linearity of a Function A function f(x) is defined linear if. f(αx 1 + βx 2 ) = αf(x 1 ) + βf(x 2 )

1. Linearity of a Function A function f(x) is defined linear if. f(αx 1 + βx 2 ) = αf(x 1 ) + βf(x 2 ) 1. Linearity of a Function A function f(x) is defined linear if f(αx 1 + βx 2 ) αf(x 1 ) + βf(x 2 ) where α and β are scalars. Example of a linear function: f(x) 2x A nonlinear function: What about f(x)

More information

6.241 Dynamic Systems and Control

6.241 Dynamic Systems and Control 6.241 Dynamic Systems and Control Lecture 7: State-space Models Readings: DDV, Chapters 7,8 Emilio Frazzoli Aeronautics and Astronautics Massachusetts Institute of Technology February 25, 2011 E. Frazzoli

More information

Principal Input and Output Directions and Hankel Singular Values

Principal Input and Output Directions and Hankel Singular Values Principal Input and Output Directions and Hankel Singular Values CEE 629 System Identification Duke University, Fall 2017 1 Continuous-time systems in the frequency domain In the frequency domain, the

More information

arxiv: v1 [cs.sy] 8 May 2015

arxiv: v1 [cs.sy] 8 May 2015 arxiv:150501958v1 cssy 8 May 2015 Direct identification of fault estimation filter for sensor faults Yiming Wan Tamas Keviczky Michel Verhaegen Delft University of Technology, 2628CD, Delft, The Netherlands

More information

4F3 - Predictive Control

4F3 - Predictive Control 4F3 Predictive Control - Lecture 2 p 1/23 4F3 - Predictive Control Lecture 2 - Unconstrained Predictive Control Jan Maciejowski jmm@engcamacuk 4F3 Predictive Control - Lecture 2 p 2/23 References Predictive

More information

Adaptive Control Using Retrospective Cost Optimization with RLS-Based Estimation for Concurrent Markov-Parameter Updating

Adaptive Control Using Retrospective Cost Optimization with RLS-Based Estimation for Concurrent Markov-Parameter Updating Joint 48th IEEE Conference on Decision and Control and 8th Chinese Control Conference Shanghai, PR China, December 6-8, 9 ThA Adaptive Control Using Retrospective Cost Optimization with RLS-Based Estimation

More information

Null Controllability of Discrete-time Linear Systems with Input and State Constraints

Null Controllability of Discrete-time Linear Systems with Input and State Constraints Proceedings of the 47th IEEE Conference on Decision and Control Cancun, Mexico, Dec. 9-11, 2008 Null Controllability of Discrete-time Linear Systems with Input and State Constraints W.P.M.H. Heemels and

More information

S deals with the problem of building a mathematical model for

S deals with the problem of building a mathematical model for JOURNAL OF GUIDANCE, CONTROL, AND DYNAMICS Vol. 18, NO. 4, July-August 1995 Identification of Linear Stochastic Systems Through Projection Filters Chung-Wen Chen* Geophysical and Environmental Research

More information

Maneuvering Target Tracking Using Retrospective-Cost Input Estimation

Maneuvering Target Tracking Using Retrospective-Cost Input Estimation I. INTRODUCTION Maneuvering Target Tracking Using Retrospective-Cost Input Estimation LIANG HAN, Student Member, IEEE ZHANG REN Beihang University Beijing, P. R. China DENNIS S. BERNSTEIN, Fellow, IEEE

More information

Module 03 Linear Systems Theory: Necessary Background

Module 03 Linear Systems Theory: Necessary Background Module 03 Linear Systems Theory: Necessary Background Ahmad F. Taha EE 5243: Introduction to Cyber-Physical Systems Email: ahmad.taha@utsa.edu Webpage: http://engineering.utsa.edu/ taha/index.html September

More information

Subspace Identification for Nonlinear Systems That are Linear in Unmeasured States 1

Subspace Identification for Nonlinear Systems That are Linear in Unmeasured States 1 Subspace Identification for Nonlinear Systems hat are Linear in nmeasured States 1 Seth L Lacy and Dennis S Bernstein 2 Abstract In this paper we apply subspace methods to the identification of a class

More information

Dynamic Attack Detection in Cyber-Physical. Systems with Side Initial State Information

Dynamic Attack Detection in Cyber-Physical. Systems with Side Initial State Information Dynamic Attack Detection in Cyber-Physical 1 Systems with Side Initial State Information Yuan Chen, Soummya Kar, and José M. F. Moura arxiv:1503.07125v1 math.oc] 24 Mar 2015 Abstract This paper studies

More information

System Theory- Based Iden2fica2on of Dynamical Models and Applica2ons

System Theory- Based Iden2fica2on of Dynamical Models and Applica2ons System Theory- Based Iden2fica2on of Dynamical Models and Applica2ons K. C. Park Center for Aerospace Structures Department of Aerospace Engineering Sciences University of Colorado at Boulder, CO, USA

More information

Linear System Theory. Wonhee Kim Lecture 1. March 7, 2018

Linear System Theory. Wonhee Kim Lecture 1. March 7, 2018 Linear System Theory Wonhee Kim Lecture 1 March 7, 2018 1 / 22 Overview Course Information Prerequisites Course Outline What is Control Engineering? Examples of Control Systems Structure of Control Systems

More information

Module 9: State Feedback Control Design Lecture Note 1

Module 9: State Feedback Control Design Lecture Note 1 Module 9: State Feedback Control Design Lecture Note 1 The design techniques described in the preceding lectures are based on the transfer function of a system. In this lecture we would discuss the state

More information

Subspace-based Identification

Subspace-based Identification of Infinite-dimensional Multivariable Systems from Frequency-response Data Department of Electrical and Electronics Engineering Anadolu University, Eskişehir, Turkey October 12, 2008 Outline 1 2 3 4 Noise-free

More information

Digital Control & Digital Filters. Lectures 13 & 14

Digital Control & Digital Filters. Lectures 13 & 14 Digital Controls & Digital Filters Lectures 13 & 14, Professor Department of Electrical and Computer Engineering Colorado State University Spring 2017 Systems with Actual Time Delays-Application 2 Case

More information

Growing Window Recursive Quadratic Optimization with Variable Regularization

Growing Window Recursive Quadratic Optimization with Variable Regularization 49th IEEE Conference on Decision and Control December 15-17, Hilton Atlanta Hotel, Atlanta, GA, USA Growing Window Recursive Quadratic Optimization with Variable Regularization Asad A. Ali 1, Jesse B.

More information

Recursive, Infinite Impulse Response (IIR) Digital Filters:

Recursive, Infinite Impulse Response (IIR) Digital Filters: Recursive, Infinite Impulse Response (IIR) Digital Filters: Filters defined by Laplace Domain transfer functions (analog devices) can be easily converted to Z domain transfer functions (digital, sampled

More information

State Estimation of Linear and Nonlinear Dynamic Systems

State Estimation of Linear and Nonlinear Dynamic Systems State Estimation of Linear and Nonlinear Dynamic Systems Part I: Linear Systems with Gaussian Noise James B. Rawlings and Fernando V. Lima Department of Chemical and Biological Engineering University of

More information

Recursive Deadbeat Controller Design

Recursive Deadbeat Controller Design NASA Technical Memorandum 112863 Recursive Deadbeat Controller Design Jer-Nan Juang Langley Research Center, Hampton, Virginia Minh Q Phan Princeton University, Princeton, New Jersey May 1997 National

More information

Retrospective Cost Adaptive Thrust Control of a 1D scramjet with Mach Number Disturbance

Retrospective Cost Adaptive Thrust Control of a 1D scramjet with Mach Number Disturbance 2015 American Control Conference Palmer House Hilton July 1-3, 2015. Chicago, IL, USA Retrospective Cost Adaptive Thrust Control of a 1D scramjet with Mach Number Disturbance Ankit Goel 1, Antai Xie 2,

More information

CHBE507 LECTURE II MPC Revisited. Professor Dae Ryook Yang

CHBE507 LECTURE II MPC Revisited. Professor Dae Ryook Yang CHBE507 LECURE II MPC Revisited Professor Dae Ryook Yang Fall 2013 Dept. of Chemical and Biological Engineering Korea University Korea University III -1 Process Models ransfer function models Fixed order

More information

Semiparametric Identification of Wiener Systems Using a Single Harmonic Input and Retrospective Cost Optimization

Semiparametric Identification of Wiener Systems Using a Single Harmonic Input and Retrospective Cost Optimization American Control Conference Marriott Waterfront, Baltimore, MD, USA June 3-July, ThC7. Semiparametric Identification of Wiener Systems Using a Single Harmonic Input and Retrospective Cost Optimization

More information

Automatic Control 2. Model reduction. Prof. Alberto Bemporad. University of Trento. Academic year

Automatic Control 2. Model reduction. Prof. Alberto Bemporad. University of Trento. Academic year Lecture: Automatic Control 2 Prof. Alberto Bemporad University of Trento Academic year 2010-2011 Prof. Alberto Bemporad (University of Trento) Automatic Control 2 Academic year 2010-2011 1 / 17 Lecture:

More information

Application of Modified Multi Model Predictive Control Algorithm to Fluid Catalytic Cracking Unit

Application of Modified Multi Model Predictive Control Algorithm to Fluid Catalytic Cracking Unit Application of Modified Multi Model Predictive Control Algorithm to Fluid Catalytic Cracking Unit Nafay H. Rehman 1, Neelam Verma 2 Student 1, Asst. Professor 2 Department of Electrical and Electronics

More information

Mathematical Induction

Mathematical Induction Lecture 6 Mathematical Induction The Mathematical Induction is a very powerful tool for proving infinitely many propositions by using only a few steps. In particular, it can be used often when the propositions

More information

MASS, STIFFNESS AND DAMPING IDENTIFICATION OF A TWO-STORY BUILDING MODEL

MASS, STIFFNESS AND DAMPING IDENTIFICATION OF A TWO-STORY BUILDING MODEL COMPDYN 2 3 rd ECCOMAS Thematic Conference on Computational Methods in Structural Dynamics and Earthquake Engineering M. Papadrakakis, M. Fragiadakis, V. Plevris (eds.) Corfu, Greece, 25-28 May 2 MASS,

More information

Extension of OKID to Output-Only System Identification

Extension of OKID to Output-Only System Identification Extension of OKID to Output-Only System Identification Francesco Vicario 1, Minh Q. Phan 2, Raimondo Betti 3, and Richard W. Longman 4 Abstract Observer/Kalman filter IDentification (OKID) is a successful

More information

Chapter 2. Real Numbers. 1. Rational Numbers

Chapter 2. Real Numbers. 1. Rational Numbers Chapter 2. Real Numbers 1. Rational Numbers A commutative ring is called a field if its nonzero elements form a group under multiplication. Let (F, +, ) be a filed with 0 as its additive identity element

More information

Problem set 5: SVD, Orthogonal projections, etc.

Problem set 5: SVD, Orthogonal projections, etc. Problem set 5: SVD, Orthogonal projections, etc. February 21, 2017 1 SVD 1. Work out again the SVD theorem done in the class: If A is a real m n matrix then here exist orthogonal matrices such that where

More information

Sequences and Summations

Sequences and Summations COMP 182 Algorithmic Thinking Sequences and Summations Luay Nakhleh Computer Science Rice University Chapter 2, Section 4 Reading Material Sequences A sequence is a function from a subset of the set of

More information

Optimal control and estimation

Optimal control and estimation Automatic Control 2 Optimal control and estimation Prof. Alberto Bemporad University of Trento Academic year 2010-2011 Prof. Alberto Bemporad (University of Trento) Automatic Control 2 Academic year 2010-2011

More information

Optimized System Identification

Optimized System Identification NASA/M-1999-209711 Optimized System Identification Jer-Nan Juang Langley Research Center, Hampton, Virginia Richard W. Longman Institute for Computer Applications in Science and Engineering (ICASE) Langley

More information

A SIMPLE METHOD FOR CONSTRUCTING ORTHOGONAL ARRAYS BY THE KRONECKER SUM

A SIMPLE METHOD FOR CONSTRUCTING ORTHOGONAL ARRAYS BY THE KRONECKER SUM Jrl Syst Sci & Complexity (2006) 19: 266 273 A SIMPLE METHOD FOR CONSTRUCTING ORTHOGONAL ARRAYS BY THE KRONECKER SUM Yingshan ZHANG Weiguo LI Shisong MAO Zhongguo ZHENG Received: 14 December 2004 / Revised:

More information

Discrete-time signals and systems

Discrete-time signals and systems Discrete-time signals and systems 1 DISCRETE-TIME DYNAMICAL SYSTEMS x(t) G y(t) Linear system: Output y(n) is a linear function of the inputs sequence: y(n) = k= h(k)x(n k) h(k): impulse response of the

More information

ORTHOGONAL ARRAYS CONSTRUCTED BY GENERALIZED KRONECKER PRODUCT

ORTHOGONAL ARRAYS CONSTRUCTED BY GENERALIZED KRONECKER PRODUCT Journal of Applied Analysis and Computation Volume 7, Number 2, May 2017, 728 744 Website:http://jaac-online.com/ DOI:10.11948/2017046 ORTHOGONAL ARRAYS CONSTRUCTED BY GENERALIZED KRONECKER PRODUCT Chun

More information

Worksheet n 6: LQG control. The case of open-cavity flow

Worksheet n 6: LQG control. The case of open-cavity flow Worksheet n 6: LQG control The case of open-cavity flow We consider the case of an open cavity at Re=6250, whose reduced-order model was obtained by the ERA algorithm for the stable subspace and an exact

More information

12.1 Arithmetic Progression Geometric Progression General things about sequences

12.1 Arithmetic Progression Geometric Progression General things about sequences ENGR11 Engineering Mathematics Lecture Notes SMS, Victoria University of Wellington Week Five. 1.1 Arithmetic Progression An arithmetic progression is a sequence where each term is found by adding a fixed

More information

Discrete-time linear systems

Discrete-time linear systems Automatic Control Discrete-time linear systems Prof. Alberto Bemporad University of Trento Academic year 2-2 Prof. Alberto Bemporad (University of Trento) Automatic Control Academic year 2-2 / 34 Introduction

More information

Properties of Open-Loop Controllers

Properties of Open-Loop Controllers Properties of Open-Loop Controllers Sven Laur University of Tarty 1 Basics of Open-Loop Controller Design Two most common tasks in controller design is regulation and signal tracking. Regulating controllers

More information

Robust Subspace System Identification via Weighted Nuclear Norm Optimization

Robust Subspace System Identification via Weighted Nuclear Norm Optimization Robust Subspace System Identification via Weighted Nuclear Norm Optimization Dorsa Sadigh Henrik Ohlsson, S. Shankar Sastry Sanjit A. Seshia University of California, Berkeley, Berkeley, CA 9470 USA. {dsadigh,ohlsson,sastry,sseshia}@eecs.berkeley.edu

More information

Shiqian Ma, MAT-258A: Numerical Optimization 1. Chapter 9. Alternating Direction Method of Multipliers

Shiqian Ma, MAT-258A: Numerical Optimization 1. Chapter 9. Alternating Direction Method of Multipliers Shiqian Ma, MAT-258A: Numerical Optimization 1 Chapter 9 Alternating Direction Method of Multipliers Shiqian Ma, MAT-258A: Numerical Optimization 2 Separable convex optimization a special case is min f(x)

More information

Module 08 Observability and State Estimator Design of Dynamical LTI Systems

Module 08 Observability and State Estimator Design of Dynamical LTI Systems Module 08 Observability and State Estimator Design of Dynamical LTI Systems Ahmad F. Taha EE 5143: Linear Systems and Control Email: ahmad.taha@utsa.edu Webpage: http://engineering.utsa.edu/ataha November

More information

State Estimation with ARMarkov Models

State Estimation with ARMarkov Models Deartment of Mechanical and Aerosace Engineering Technical Reort No. 3046, October 1998. Princeton University, Princeton, NJ. State Estimation with ARMarkov Models Ryoung K. Lim 1 Columbia University,

More information

IMPULSIVE CONTROL OF DISCRETE-TIME NETWORKED SYSTEMS WITH COMMUNICATION DELAYS. Shumei Mu, Tianguang Chu, and Long Wang

IMPULSIVE CONTROL OF DISCRETE-TIME NETWORKED SYSTEMS WITH COMMUNICATION DELAYS. Shumei Mu, Tianguang Chu, and Long Wang IMPULSIVE CONTROL OF DISCRETE-TIME NETWORKED SYSTEMS WITH COMMUNICATION DELAYS Shumei Mu Tianguang Chu and Long Wang Intelligent Control Laboratory Center for Systems and Control Department of Mechanics

More information

Reduced-order models for control of fluids using the eigensystem realization algorithm

Reduced-order models for control of fluids using the eigensystem realization algorithm Theor. Comput. Fluid Dyn. DOI 0.007/s0062-00-084-8 ORIGINAL ARTICLE Zhanhua Ma Sunil Ahuja Clarence W. Rowley Reduced-order models for control of fluids using the eigensystem realization algorithm Received:

More information

CDS Solutions to Final Exam

CDS Solutions to Final Exam CDS 22 - Solutions to Final Exam Instructor: Danielle C Tarraf Fall 27 Problem (a) We will compute the H 2 norm of G using state-space methods (see Section 26 in DFT) We begin by finding a minimal state-space

More information

EEE582 Homework Problems

EEE582 Homework Problems EEE582 Homework Problems HW. Write a state-space realization of the linearized model for the cruise control system around speeds v = 4 (Section.3, http://tsakalis.faculty.asu.edu/notes/models.pdf). Use

More information

Minimum modelling retrospective cost adaptive control of uncertain Hammerstein systems using auxiliary nonlinearities

Minimum modelling retrospective cost adaptive control of uncertain Hammerstein systems using auxiliary nonlinearities This article was downloaded by: [University of Michigan] On: 02 February 2014, At: 02:26 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office:

More information

Extended Kalman Filter based State Estimation of Wind Turbine

Extended Kalman Filter based State Estimation of Wind Turbine Extended Kalman Filter based State Estimation of Wind Turbine Kavitha N*, Vijayachitra S# *PG Scholar,.E. Control and Instrumentation Engineering #Professor, Department of Electronics and Instrumentation

More information

REACHABILITY AND OBSERVABILITY OF LINEAR SYSTEMS OVER MAX-PLUS 1

REACHABILITY AND OBSERVABILITY OF LINEAR SYSTEMS OVER MAX-PLUS 1 KYBERNETIKA VOLUME 35 (1999), NUMBER 1, PAGES 2-12 REACHABILITY AND OBSERVABILITY OF LINEAR SYSTEMS OVER MAX-PLUS 1 MICHAEL J. GAZARIK AND EDWARD W. KAMEN This paper discusses the properties of reachability

More information

Lecture 5: Recurrent Neural Networks

Lecture 5: Recurrent Neural Networks 1/25 Lecture 5: Recurrent Neural Networks Nima Mohajerin University of Waterloo WAVE Lab nima.mohajerin@uwaterloo.ca July 4, 2017 2/25 Overview 1 Recap 2 RNN Architectures for Learning Long Term Dependencies

More information

Outline. 1 Linear Quadratic Problem. 2 Constraints. 3 Dynamic Programming Solution. 4 The Infinite Horizon LQ Problem.

Outline. 1 Linear Quadratic Problem. 2 Constraints. 3 Dynamic Programming Solution. 4 The Infinite Horizon LQ Problem. Model Predictive Control Short Course Regulation James B. Rawlings Michael J. Risbeck Nishith R. Patel Department of Chemical and Biological Engineering Copyright c 217 by James B. Rawlings Outline 1 Linear

More information

Sliding Window Recursive Quadratic Optimization with Variable Regularization

Sliding Window Recursive Quadratic Optimization with Variable Regularization 11 American Control Conference on O'Farrell Street, San Francisco, CA, USA June 29 - July 1, 11 Sliding Window Recursive Quadratic Optimization with Variable Regularization Jesse B. Hoagg, Asad A. Ali,

More information

Linear Systems. Manfred Morari Melanie Zeilinger. Institut für Automatik, ETH Zürich Institute for Dynamic Systems and Control, ETH Zürich

Linear Systems. Manfred Morari Melanie Zeilinger. Institut für Automatik, ETH Zürich Institute for Dynamic Systems and Control, ETH Zürich Linear Systems Manfred Morari Melanie Zeilinger Institut für Automatik, ETH Zürich Institute for Dynamic Systems and Control, ETH Zürich Spring Semester 2016 Linear Systems M. Morari, M. Zeilinger - Spring

More information

GENERALIZATION OF UNIVERSAL PARTITION AND BIPARTITION THEOREMS. Hacène Belbachir 1

GENERALIZATION OF UNIVERSAL PARTITION AND BIPARTITION THEOREMS. Hacène Belbachir 1 #A59 INTEGERS 3 (23) GENERALIZATION OF UNIVERSAL PARTITION AND BIPARTITION THEOREMS Hacène Belbachir USTHB, Faculty of Mathematics, RECITS Laboratory, Algiers, Algeria hbelbachir@usthb.dz, hacenebelbachir@gmail.com

More information

4F3 - Predictive Control

4F3 - Predictive Control 4F3 Predictive Control - Discrete-time systems p. 1/30 4F3 - Predictive Control Discrete-time State Space Control Theory For reference only Jan Maciejowski jmm@eng.cam.ac.uk 4F3 Predictive Control - Discrete-time

More information

The Impulse Signal. A signal I(k), which is defined as zero everywhere except at a single point, where its value is equal to 1

The Impulse Signal. A signal I(k), which is defined as zero everywhere except at a single point, where its value is equal to 1 The Impulse Signal A signal I(k), which is defined as zero everywhere except at a single point, where its value is equal to 1 I = [0 0 1 0 0 0]; stem(i) Filter Impulse through Feedforward Filter 0 0.25.5.25

More information

Adaptive Control of an Aircraft with Uncertain Nonminimum-Phase Dynamics

Adaptive Control of an Aircraft with Uncertain Nonminimum-Phase Dynamics 1 American Control Conference Palmer House Hilton July 1-3, 1. Chicago, IL, USA Adaptive Control of an Aircraft with Uncertain Nonminimum-Phase Dynamics Ahmad Ansari and Dennis S. Bernstein Abstract This

More information

SINGLE DEGREE OF FREEDOM SYSTEM IDENTIFICATION USING LEAST SQUARES, SUBSPACE AND ERA-OKID IDENTIFICATION ALGORITHMS

SINGLE DEGREE OF FREEDOM SYSTEM IDENTIFICATION USING LEAST SQUARES, SUBSPACE AND ERA-OKID IDENTIFICATION ALGORITHMS 3 th World Conference on Earthquake Engineering Vancouver, B.C., Canada August -6, 24 Paper No. 278 SINGLE DEGREE OF FREEDOM SYSTEM IDENTIFICATION USING LEAST SQUARES, SUBSPACE AND ERA-OKID IDENTIFICATION

More information

Eigenvalue Constraints for Realization-Based Identification. University of California San Diego, La Jolla, CA, USA. Martin J.

Eigenvalue Constraints for Realization-Based Identification. University of California San Diego, La Jolla, CA, USA. Martin J. Proc AIAA Atmospheric Flight Mechanics Conference AIAA-2012-4951, 16 pages, Minneapolis, MN, USA (2012) DOI: 102514/62012-4951 Eigenvalue Constraints for Realization-Based Identification Daniel N Miller

More information

State Estimation using Moving Horizon Estimation and Particle Filtering

State Estimation using Moving Horizon Estimation and Particle Filtering State Estimation using Moving Horizon Estimation and Particle Filtering James B. Rawlings Department of Chemical and Biological Engineering UW Math Probability Seminar Spring 2009 Rawlings MHE & PF 1 /

More information

RELATIVE NAVIGATION FOR SATELLITES IN CLOSE PROXIMITY USING ANGLES-ONLY OBSERVATIONS

RELATIVE NAVIGATION FOR SATELLITES IN CLOSE PROXIMITY USING ANGLES-ONLY OBSERVATIONS (Preprint) AAS 12-202 RELATIVE NAVIGATION FOR SATELLITES IN CLOSE PROXIMITY USING ANGLES-ONLY OBSERVATIONS Hemanshu Patel 1, T. Alan Lovell 2, Ryan Russell 3, Andrew Sinclair 4 "Relative navigation using

More information

Zero controllability in discrete-time structured systems

Zero controllability in discrete-time structured systems 1 Zero controllability in discrete-time structured systems Jacob van der Woude arxiv:173.8394v1 [math.oc] 24 Mar 217 Abstract In this paper we consider complex dynamical networks modeled by means of state

More information

ADAPTIVE FILTER ALGORITHMS. Prepared by Deepa.T, Asst.Prof. /TCE

ADAPTIVE FILTER ALGORITHMS. Prepared by Deepa.T, Asst.Prof. /TCE ADAPTIVE FILTER ALGORITHMS Prepared by Deepa.T, Asst.Prof. /TCE Equalization Techniques Fig.3 Classification of equalizers Equalizer Techniques Linear transversal equalizer (LTE, made up of tapped delay

More information

Discrete-time models and control

Discrete-time models and control Discrete-time models and control Silvano Balemi University of Applied Sciences of Southern Switzerland Zürich, 2009-2010 Discrete-time signals 1 Step response of a sampled system Sample and hold 2 Sampling

More information

Outline. Linear regulation and state estimation (LQR and LQE) Linear differential equations. Discrete time linear difference equations

Outline. Linear regulation and state estimation (LQR and LQE) Linear differential equations. Discrete time linear difference equations Outline Linear regulation and state estimation (LQR and LQE) James B. Rawlings Department of Chemical and Biological Engineering 1 Linear Quadratic Regulator Constraints The Infinite Horizon LQ Problem

More information

System Identification and Models for Flight Control

System Identification and Models for Flight Control System Identification and Models for Flight Control -./,1231-44567 &! %! $! #! "! x α α k+1 A ERA x = 1 t α 1 α x C L (k t) = C ERA C Lα C L α α α ERA Model k + quasi-steady contribution k B ERA t α k

More information

UNIT V FINITE WORD LENGTH EFFECTS IN DIGITAL FILTERS PART A 1. Define 1 s complement form? In 1,s complement form the positive number is represented as in the sign magnitude form. To obtain the negative

More information

Washout Filters in Feedback Control: Benefits, Limitations and Extensions

Washout Filters in Feedback Control: Benefits, Limitations and Extensions Washout Filters in Feedback Control: Benefits, Limitations and Extensions Munther A. Hassouneh 1, Hsien-Chiarn Lee 2 and Eyad H. Abed 1 1 Department of Electrical and Computer Engineering and the Institute

More information

Subspace Identification With Guaranteed Stability Using Constrained Optimization

Subspace Identification With Guaranteed Stability Using Constrained Optimization IEEE TANSACTIONS ON AUTOMATIC CONTOL, VOL. 48, NO. 7, JULY 2003 259 Subspace Identification With Guaranteed Stability Using Constrained Optimization Seth L. Lacy and Dennis S. Bernstein Abstract In system

More information

Introduction to Binary Convolutional Codes [1]

Introduction to Binary Convolutional Codes [1] Introduction to Binary Convolutional Codes [1] Yunghsiang S. Han Graduate Institute of Communication Engineering, National Taipei University Taiwan E-mail: yshan@mail.ntpu.edu.tw Y. S. Han Introduction

More information

Comparison of four state observer design algorithms for MIMO system

Comparison of four state observer design algorithms for MIMO system Archives of Control Sciences Volume 23(LIX), 2013 No. 2, pages 131 144 Comparison of four state observer design algorithms for MIMO system VINODH KUMAR. E, JOVITHA JEROME and S. AYYAPPAN A state observer

More information

A METHOD FOR SOLVING DUAL FUZZY GENERAL LINEAR SYSTEMS*

A METHOD FOR SOLVING DUAL FUZZY GENERAL LINEAR SYSTEMS* Appl Comput Math 7 (2008) no2 pp235-241 A METHOD FOR SOLVING DUAL FUZZY GENERAL LINEAR SYSTEMS* REZA EZZATI Abstract In this paper the main aim is to develop a method for solving an arbitrary m n dual

More information

1 Similarity transform 2. 2 Controllability The PBH test for controllability Observability The PBH test for observability...

1 Similarity transform 2. 2 Controllability The PBH test for controllability Observability The PBH test for observability... Contents 1 Similarity transform 2 2 Controllability 3 21 The PBH test for controllability 5 3 Observability 6 31 The PBH test for observability 7 4 Example ([1, pp121) 9 5 Subspace decomposition 11 51

More information

Interchange of Filtering and Downsampling/Upsampling

Interchange of Filtering and Downsampling/Upsampling Interchange of Filtering and Downsampling/Upsampling Downsampling and upsampling are linear systems, but not LTI systems. They cannot be implemented by difference equations, and so we cannot apply z-transform

More information

Advanced Digital Signal Processing -Introduction

Advanced Digital Signal Processing -Introduction Advanced Digital Signal Processing -Introduction LECTURE-2 1 AP9211- ADVANCED DIGITAL SIGNAL PROCESSING UNIT I DISCRETE RANDOM SIGNAL PROCESSING Discrete Random Processes- Ensemble Averages, Stationary

More information

Model Predictive Control Short Course Regulation

Model Predictive Control Short Course Regulation Model Predictive Control Short Course Regulation James B. Rawlings Michael J. Risbeck Nishith R. Patel Department of Chemical and Biological Engineering Copyright c 2017 by James B. Rawlings Milwaukee,

More information

On Input Design for System Identification

On Input Design for System Identification On Input Design for System Identification Input Design Using Markov Chains CHIARA BRIGHENTI Masters Degree Project Stockholm, Sweden March 2009 XR-EE-RT 2009:002 Abstract When system identification methods

More information