A Critique of Observers Used in the Context of Feedback Control

One of the core tenets of feedback control is that a system’s state contains all of the information necessary to predict a system’s future response given future inputs. If the state is not directly measured then it can be estimated using a suitably designed observer. This is a powerful idea with widespread consequences. This paper will present a critique of the use of observers in feedback control. Benefits and drawbacks will be highlighted including fundamental design limitations. The analysis will be illustrated by several real world examples including robots executing a repetitive task, relay autotuning in the presence of broadband disturbances, power line signalling in AC microgrid power systems, Type 1 diabetes management and harmonic suppression in power electronics.

[1]  Graham C. Goodwin,et al.  Design of modulated and demodulated controllers for flexible structures , 2007 .

[2]  Alan E. Gelfand,et al.  Bayesian statistics without tears: A sampling-resampling perspective , 1992 .

[3]  William S. Levine,et al.  The Control Handbook , 2005 .

[4]  Eyal Dassau,et al.  Coordinated Basal—Bolus Infusion for Tighter Postprandial Glucose Control in Insulin Pump Therapy , 2009, Journal of diabetes science and technology.

[5]  Graham C. Goodwin,et al.  A fundamental control performance limit for a class of positive nonlinear systems , 2018, Autom..

[6]  M. Atkinson,et al.  Type 1 diabetes , 2014, The Lancet.

[7]  F. Doyle,et al.  Quest for the Artificial Pancreas: Combining Technology with Treatment , 2010, IEEE Engineering in Medicine and Biology Magazine.

[8]  Hassan K. Khalil,et al.  Robust stabilization of non-minimum phase nonlinear systems using extended high gain observers , 2011, 2008 American Control Conference.

[9]  Eyal Dassau,et al.  Periodic zone-MPC with asymmetric costs for outpatient-ready safety of an artificial pancreas to treat type 1 diabetes , 2016, Autom..

[10]  K. Kumareswaran.  Closed-loop insulin delivery in adults with type 1 diabetes , 2012 .

[11]  Mohammed M'Saad,et al.  High gain observer for a class of non-triangular systems , 2011, Syst. Control. Lett..

[12]  Graham C. Goodwin,et al.  Connecting filtering and control sensitivity functions , 2014, Autom..

[13]  David Q. Mayne,et al.  Constrained state estimation for nonlinear discrete-time systems: stability and moving horizon approximations , 2003, IEEE Trans. Autom. Control..

[14]  Graham C. Goodwin,et al.  Properties of modulated and demodulated systems with implications to feedback limitations , 2005, Autom..

[15]  D. Luenberger.  Observers for multivariable systems , 1966 .

[16]  Tore Hägglund,et al.  Asymmetric relay autotuning - Practical features for industrial use , 2016 .

[17]  Hassan K. Khalil,et al.  Lyapunov-based switching control of nonlinear systems using high-gain observers , 2005, Proceedings of the 2005, American Control Conference, 2005..

[18]  G. Stein,et al.  Respect the unstable , 2003 .

[19]  H.F. Durrant-Whyte,et al.  A new approach for filtering nonlinear systems , 1995, Proceedings of 1995 American Control Conference - ACC'95.

[20]  Mohamed Boutayeb,et al.  A strong tracking extended Kalman observer for nonlinear discrete-time systems , 1999, IEEE Trans. Autom. Control..

[21]  Arthur J. Krener,et al.  Linearization by output injection and nonlinear observers , 1983 .

[22]  Graham C. Goodwin,et al.  Harmonic suppression and delay compensation for inverters via variable horizon nonlinear model predictive control , 2015, Int. J. Control.

[23]  P. Khargonekar,et al.  State-space solutions to standard H/sub 2/ and H/sub infinity / control problems , 1989 .

[24]  S. Drakunov.  Adaptive quasioptimal filter with discontinuous parameters , 1984 .

[25]  J. Gauthier,et al.  High gain estimation for nonlinear systems , 1992 .

[26]  A. Jazwinski.  Stochastic Processes and Filtering Theory , 1970 .

[27]  Graham C. Goodwin,et al.  Lagrangian duality between constrained estimation and control , 2005, Autom..

[28]  Garry M. Steil,et al.  Identification of Intraday Metabolic Profiles during Closed-Loop Glucose Control in Individuals with Type 1 Diabetes , 2009, Journal of diabetes science and technology.

[29]  Josep Vehí,et al.  A review of personalized blood glucose prediction strategies for T1DM patients , 2017, International journal for numerical methods in biomedical engineering.

[30]  R. E. Kalman,et al.  A New Approach to Linear Filtering and Prediction Problems , 2002 .

[31]  Neil J. Gordon,et al.  A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking , 2002, IEEE Trans. Signal Process..

[32]  G. Mirzaeva,et al.  Use of harmonic power line communication to enhance a decentralised control method of parallel inverters in an AC microgrid , 2017, 2017 IEEE Southern Power Electronics Conference (SPEC).

[33]  G. Ellis.  Observers in Control Systems: A Practical Guide , 2002 .

[34]  Kumpati S. Narendra,et al.  A new approach to adaptive control using multiple models , 2012 .

[35]  Graham C. Goodwin,et al.  Identifiability of errors in variables dynamic systems , 2008, Autom..

[36]  S. Aronoff,et al.  Glucose Metabolism and Regulation: Beyond Insulin and Glucagon , 2004 .

[37]  Roman Hovorka,et al.  The future of continuous glucose monitoring: closed loop. , 2008, Current diabetes reviews.

[38]  R. E. Kalman,et al.  New Results in Linear Filtering and Prediction Theory , 1961 .

[39]  G. Goodwin,et al.  Sampling and Sampled-Data Models: The Interface Between the Continuous World and Digital Algorithms , 2013, IEEE Control Systems.

[40]  Daniel Howsmon,et al.  Closed-Loop Control Without Meal Announcement in Type 1 Diabetes. , 2017, Diabetes technology & therapeutics.

[41]  G. Zames.  Feedback and optimal sensitivity: Model reference transformations, multiplicative seminorms, and approximate inverses , 1981 .

[42]  Simon J. Godsill,et al.  On sequential Monte Carlo sampling methods for Bayesian filtering , 2000, Stat. Comput..

[43]  R. Bergman.  Minimal Model: Perspective from 2005 , 2006, Hormone Research in Paediatrics.

[44]  L. Magni,et al.  Multinational Study of Subcutaneous Model-Predictive Closed-Loop Control in Type 1 Diabetes Mellitus: Summary of the Results , 2010, Journal of diabetes science and technology.

[45]  Hassan K. Khalil,et al.  Closed-Loop Behavior of a Class of Nonlinear Systems Under EKF-Based Control , 2007, IEEE Transactions on Automatic Control.

[46]  D. Mayne,et al.  Monte Carlo techniques to estimate the conditional expectation in multi-stage non-linear filtering† , 1969 .

[47]  Zhiqiang Gao,et al.  A survey of state and disturbance observers for practitioners , 2006, 2006 American Control Conference.

[48]  M. H. Mickle,et al.  Disturbance estimation and compensation in linear systems , 1990 .

[49]  Ali Saberi,et al.  The discrete‐time H∞ control problem with measurement feedback , 1994 .

[50]  G. Stein,et al.  Multivariable feedback design: Concepts for a classical/modern synthesis , 1981 .

[51]  Peter J. Thomas,et al.  Roll Eccentricity Control for Strip Rolling Mills , 1987 .

[52]  Robert F. Stengel,et al.  Optimal Control and Estimation , 1994 .

[53]  J. Grizzle,et al.  The Extended Kalman Filter as a Local Asymptotic Observer for Nonlinear Discrete-Time Systems , 1992, 1992 American Control Conference.

[54]  Vadim I. Utkin,et al.  Sliding mode control in electromechanical systems , 1999 .

[55]  J. Leahy,et al.  Fully Automated Closed-Loop Insulin Delivery Versus Semiautomated Hybrid Control in Pediatric Patients With Type 1 Diabetes Using an Artificial Pancreas , 2008 .

[56]  Graham C. Goodwin,et al.  A fundamental control limitation for linear positive systems with application to Type 1 diabetes treatment , 2015, Autom..

[57]  W. Tamborlane,et al.  The Artificial Pancreas: Are We There Yet? , 2014, Diabetes Care.

[58]  Konrad Reif,et al.  An EKF-Based Nonlinear Observer with a Prescribed Degree of Stability , 1998, Autom..

[59]  B. Wayne Bequette,et al.  Challenges and recent progress in the development of a closed-loop artificial pancreas , 2012, Annu. Rev. Control..

[60]  A. Germani,et al.  A Luenberger-like observer for nonlinear systems , 1993 .

[61]  Graham C. Goodwin,et al.  A modified relay autotuner for systems having large broadband disturbances , 2018, Autom..

[62]  M. Morari,et al.  Closed-Loop Control of Blood Glucose , 2007 .

[63]  Hassan K. Khalil,et al.  High-gain observers in nonlinear feedback control , 2009, 2009 IEEE International Conference on Control and Automation.

[64]  Gregory P. Forlenza,et al.  Optimizing Hybrid Closed-Loop Therapy in Adolescents and Emerging Adults Using the MiniMed 670G System , 2018, Diabetes Care.

[65]  Tore Hägglund,et al.  Automatic tuning of simple regulators with specifications on phase and amplitude margins , 1984, Autom..

[66]  G. Goodwin,et al.  High-speed digital signal processing and control , 1992, Proc. IEEE.

[67]  Jakub Bernat,et al.  Multi-modelling as new estimation schema for high-gain observers , 2015, Int. J. Control.

[68]  N. Gordon,et al.  Novel approach to nonlinear/non-Gaussian Bayesian state estimation , 1993 .

[69]  Hassan K. Khalil,et al.  Full-order high-gain observers for minimum phase nonlinear systems , 2009, Proceedings of the 48h IEEE Conference on Decision and Control (CDC) held jointly with 2009 28th Chinese Control Conference.

[70]  A. Krener.  The Convergence of the Extended Kalman Filter , 2002, math/0212255.

[71]  Ali Saberi,et al.  The discrete-time $H_\infty$ control problem with strictly proper measurement feedback , 1993 .

[72]  Lauren M. Huyett,et al.  Closed-Loop Artificial Pancreas Systems: Engineering the Algorithms , 2014, Diabetes Care.

[73]  Robin J. Evans,et al.  Sinusoidal disturbance rejection with application to helicopter flight data estimation , 1986, IEEE Trans. Acoust. Speech Signal Process..

[74]  Hassan Hammouri,et al.  A new procedure for time-varying linearization up to output injection , 1996 .

[75]  C. Cobelli,et al.  Progress in Development of an Artificial Pancreas , 2009, Journal of diabetes science and technology.

[76]  S. Drakunov.  Sliding-mode observers based on equivalent control method , 1992, [1992] Proceedings of the 31st IEEE Conference on Decision and Control.

[77]  F. Daum.  Nonlinear filters: beyond the Kalman filter , 2005, IEEE Aerospace and Electronic Systems Magazine.

[78]  B. Bequette.  A critical assessment of algorithms and challenges in the development of a closed-loop artificial pancreas. , 2005, Diabetes technology & therapeutics.

[79]  C. Cobelli,et al.  The Artificial Pancreas in 2016: A Digital Treatment Ecosystem for Diabetes , 2016, Diabetes Care.

[80]  Zhiqiang Gao,et al.  A comparison study of advanced state observer design techniques , 2003, Proceedings of the 2003 American Control Conference, 2003..

[81]  Michael Green,et al.  Discrete-time H/sup infinity / control , 1989, Proceedings of the 28th IEEE Conference on Decision and Control,.

[82]  Graham C. Goodwin,et al.  A Method for Improving the Dynamic Accuracy of a Robot Performing a Repetitive Task , 1989, Int. J. Robotics Res..

[83]  Graham C. Goodwin,et al.  A performance bound for optimal insulin infusion in individuals with Type 1 diabetes ingesting a meal with slow postprandial response , 2019, Autom..

[84]  Josep Vehí,et al.  Unannounced Meals in the Artificial Pancreas: Detection Using Continuous Glucose Monitoring , 2018, Sensors.

[85]  D. Luenberger.  Observing the State of a Linear System , 1964, IEEE Transactions on Military Electronics.

[86]  Torsten Söderström,et al.  Errors-in-variables methods in system identification , 2018, Autom..

[87]  James B. Rawlings,et al.  Particle filtering and moving horizon estimation , 2006, Comput. Chem. Eng..

[88]  Graham C. Goodwin,et al.  Adaptive filtering prediction and control , 1984 .