Solutions to the inverse LQR problem with application to biological systems analysis

M. Cody Priess, Richard Conway, Jongeun Choi, John M. Popovich, Clark Radcliffe

Research output: Contribution to journalArticle

34 Citations (Scopus)

Abstract

In this brief, we present a set of techniques for finding a cost function to the time-invariant linear quadratic regulator (LQR) problem in both continuous- and discrete-time cases. Our methodology is based on the solution to the inverse LQR problem, which can be stated as: does a given controller K describe the solution to a time-invariant LQR problem, and if so, what weights Q and R produce K as the optimal solution? Our motivation for investigating this problem is the analysis of motion goals in biological systems. We first describe an efficient linear matrix inequality (LMI) method for determining a solution to the general case of this inverse LQR problem when both the weighting matrices Q and R are unknown. Our first LMI-based formulation provides a unique solution when it is feasible. In addition, we propose a gradient-based, least-squares minimization method that can be applied to approximate a solution in cases when the LMIs are infeasible. This new method is very useful in practice since the estimated gain matrix K from the noisy experimental data could be perturbed by the estimation error, which may result in the infeasibility of the LMIs. We also provide an LMI minimization problem to find a good initial point for the minimization using the proposed gradient descent algorithm. We then provide a set of examples to illustrate how to apply our approaches to several different types of problems. An important result is the application of the technique to human subject posture control when seated on a moving robot. Results show that we can recover a cost function which may provide a useful insight on the human motor control goal.

Original languageEnglish
Article number6880317
Pages (from-to)770-777
Number of pages8
JournalIEEE Transactions on Control Systems Technology
Volume23
Issue number2
DOIs
Publication statusPublished - 2015 Mar 1

Fingerprint

Biological systems
Linear matrix inequalities
Systems analysis
Cost functions
Error analysis
Robots
Controllers

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Cite this

Priess, M. Cody ; Conway, Richard ; Choi, Jongeun ; Popovich, John M. ; Radcliffe, Clark. / Solutions to the inverse LQR problem with application to biological systems analysis. In: IEEE Transactions on Control Systems Technology. 2015 ; Vol. 23, No. 2. pp. 770-777.
@article{fbdc713fc914401c935746220b63db7e,
title = "Solutions to the inverse LQR problem with application to biological systems analysis",
abstract = "In this brief, we present a set of techniques for finding a cost function to the time-invariant linear quadratic regulator (LQR) problem in both continuous- and discrete-time cases. Our methodology is based on the solution to the inverse LQR problem, which can be stated as: does a given controller K describe the solution to a time-invariant LQR problem, and if so, what weights Q and R produce K as the optimal solution? Our motivation for investigating this problem is the analysis of motion goals in biological systems. We first describe an efficient linear matrix inequality (LMI) method for determining a solution to the general case of this inverse LQR problem when both the weighting matrices Q and R are unknown. Our first LMI-based formulation provides a unique solution when it is feasible. In addition, we propose a gradient-based, least-squares minimization method that can be applied to approximate a solution in cases when the LMIs are infeasible. This new method is very useful in practice since the estimated gain matrix K from the noisy experimental data could be perturbed by the estimation error, which may result in the infeasibility of the LMIs. We also provide an LMI minimization problem to find a good initial point for the minimization using the proposed gradient descent algorithm. We then provide a set of examples to illustrate how to apply our approaches to several different types of problems. An important result is the application of the technique to human subject posture control when seated on a moving robot. Results show that we can recover a cost function which may provide a useful insight on the human motor control goal.",
author = "Priess, {M. Cody} and Richard Conway and Jongeun Choi and Popovich, {John M.} and Clark Radcliffe",
year = "2015",
month = "3",
day = "1",
doi = "10.1109/TCST.2014.2343935",
language = "English",
volume = "23",
pages = "770--777",
journal = "IEEE Transactions on Control Systems Technology",
issn = "1063-6536",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "2",

}

Solutions to the inverse LQR problem with application to biological systems analysis. / Priess, M. Cody; Conway, Richard; Choi, Jongeun; Popovich, John M.; Radcliffe, Clark.

In: IEEE Transactions on Control Systems Technology, Vol. 23, No. 2, 6880317, 01.03.2015, p. 770-777.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Solutions to the inverse LQR problem with application to biological systems analysis

AU - Priess, M. Cody

AU - Conway, Richard

AU - Choi, Jongeun

AU - Popovich, John M.

AU - Radcliffe, Clark

PY - 2015/3/1

Y1 - 2015/3/1

N2 - In this brief, we present a set of techniques for finding a cost function to the time-invariant linear quadratic regulator (LQR) problem in both continuous- and discrete-time cases. Our methodology is based on the solution to the inverse LQR problem, which can be stated as: does a given controller K describe the solution to a time-invariant LQR problem, and if so, what weights Q and R produce K as the optimal solution? Our motivation for investigating this problem is the analysis of motion goals in biological systems. We first describe an efficient linear matrix inequality (LMI) method for determining a solution to the general case of this inverse LQR problem when both the weighting matrices Q and R are unknown. Our first LMI-based formulation provides a unique solution when it is feasible. In addition, we propose a gradient-based, least-squares minimization method that can be applied to approximate a solution in cases when the LMIs are infeasible. This new method is very useful in practice since the estimated gain matrix K from the noisy experimental data could be perturbed by the estimation error, which may result in the infeasibility of the LMIs. We also provide an LMI minimization problem to find a good initial point for the minimization using the proposed gradient descent algorithm. We then provide a set of examples to illustrate how to apply our approaches to several different types of problems. An important result is the application of the technique to human subject posture control when seated on a moving robot. Results show that we can recover a cost function which may provide a useful insight on the human motor control goal.

AB - In this brief, we present a set of techniques for finding a cost function to the time-invariant linear quadratic regulator (LQR) problem in both continuous- and discrete-time cases. Our methodology is based on the solution to the inverse LQR problem, which can be stated as: does a given controller K describe the solution to a time-invariant LQR problem, and if so, what weights Q and R produce K as the optimal solution? Our motivation for investigating this problem is the analysis of motion goals in biological systems. We first describe an efficient linear matrix inequality (LMI) method for determining a solution to the general case of this inverse LQR problem when both the weighting matrices Q and R are unknown. Our first LMI-based formulation provides a unique solution when it is feasible. In addition, we propose a gradient-based, least-squares minimization method that can be applied to approximate a solution in cases when the LMIs are infeasible. This new method is very useful in practice since the estimated gain matrix K from the noisy experimental data could be perturbed by the estimation error, which may result in the infeasibility of the LMIs. We also provide an LMI minimization problem to find a good initial point for the minimization using the proposed gradient descent algorithm. We then provide a set of examples to illustrate how to apply our approaches to several different types of problems. An important result is the application of the technique to human subject posture control when seated on a moving robot. Results show that we can recover a cost function which may provide a useful insight on the human motor control goal.

UR - http://www.scopus.com/inward/record.url?scp=85027957809&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85027957809&partnerID=8YFLogxK

U2 - 10.1109/TCST.2014.2343935

DO - 10.1109/TCST.2014.2343935

M3 - Article

AN - SCOPUS:85027957809

VL - 23

SP - 770

EP - 777

JO - IEEE Transactions on Control Systems Technology

JF - IEEE Transactions on Control Systems Technology

SN - 1063-6536

IS - 2

M1 - 6880317

ER -