This book introduces robust estimation and failure detection, with a thorough presentation of Kalman filtering and H-infinity filtering theory. These estimation techniques make it possible for engineers to design estimators that are more general and robust. The book also reviews the likelihood ratio method for failure detection and demonstrates how to design failure detectors that are sensitive to failures but insensitive to model variations. This book will give engineers a concise presentation of these important techniques, as well as an overview of important robust control developments of the last fifteen years.

Inhalt

1 Introduction.- 2 Estimation and Failure Detection: An Overview.- 2.1 Introduction.- 2.2 Ever Since Wiener.- 2.2.1 Wiener and Kalman Filters.- 2.2.2 Beyond Linear Least Squares Estimation.- 2.2.3 Kalman Filter and Model Uncertainties.- 2.2.4 Robust Estimation and the Small Gain Theorem.- 2.2.5 Robust Stability and Robust Performance for Estimation.- 2.2.6 Further Discussion on Robust Estimation and Control.- 2.2.7 Risk Sensitive Control and Estimation.- 2.3 Failure Detection and Isolation.- 2.3.1 Kalman Filters in FDI Algorithms: The GLUT.- 2.3.2 Nonadditive Failures.- 2.3.3 Modeling Uncertainties and FDI Algorithms.- 2.3.4 Robust Failure Detection and Isolation.- 2.4 Summary.- 3 Discrete-Time Robust Estimation.- 3.1 Introduction.- 3.2 Plants with an Uncertain Noise Model.- 3.2.1 Problem Formulation.- 3.2.2 Derivation of the Estimator.- 3.2.3 Estimator Properties.- 3.2.4 Estimator Equations and Discussion.- 3.3 Plants with Uncertain Dynamics and Noise Model.- 3.3.1 Problem Formulation.- 3.3.2 Derivation of the Estimator.- 3.3.3 Robust Estimator Equations and Discussion.- 3.4 Extension to Steady State.- 3.5 Robust Fixed-Interval Smoothing.- 3.6 Numerical Examples.- 3.6.1 A Two-State System.- 3.6.2 Attitude Determination.- 3.7 Related Work.- 4 Stochastic Interpretation of Robust Estimation: Risk Sensitivity.- 4.1 Introduction.- 4.2 The Risk Sensitive Optimal Estimation Problem.- 4.2.1 Problem Formulation.- 4.2.2 Equivalence to Game Theoretic Estimation.- 4.3 Extension to Systems with Modeling Uncertainty.- 4.4 Numerical Comparison of Error Density Functions.- 4.5 Summary.- 5 Robust Failure Detection and Isolation.- 5.1 Introduction.- 5.2 Problem Description.- 5.2.1 General Discussion and Notation.- 5.2.2 Problem Formulation.- 5.2.3 The Failure Model.- 5.3 A Likelihood Ratio Test with Robustness Properties.- 5.4 Likelihood Ratio Tests and Plant Uncertainties.- 5.4.1 Examples: Underwater Vehicle with Model Uncertainty.- 5.5 FDI with Robustness to Failure Mode, Noise and Plant Uncertainties.- 5.5.1 The Decision Function.- 5.5.2 Robust Estimator Design.- 5.5.3 Summary of the Algorithm.- 5.6 Summary.- 6 Two Applications.- 6.1 Introduction.- 6.2 Application to an Underwater Vehicle.- 6.2.1 Straight and Level Cruise.- 6.2.2 Maneuvers.- 6.3 Application to Reentry Vehicle attitude Control Systems.- 6.3.1 Problem Description.- 6.3.2 Robust FDI Filter Architecture.- 6.3.3 Results.- 6.3.4 Summary.- A The Kalman Filter.- A.1 Problem Description.- A.2 The One-Step Predictor.- A.3 Measurement Update and the Filtered Estimate.- A.4 Gaussian Disturbance.- A.5 The Innovation Process.- A.6 Linear Time-Invariant Systems.- A.7 The Wiener Filter.- A.8 Smoothing.- A.8.1 Fixed-Interval Smoothing.- A.8.2 Fixed-Point and Fixed-Lag Smoothing.- A.9 The Extended Kaiman Filter (EKF).- A.10 Summary of Equations and Additional Remarks.- B Outputs of Linear Systems and Their Quadratic Forms.- B.1 Moments of Linear Systems Outputs.- B.2 Probability Density Functions of Gaussian Quadratic Forms.- C Continuous-Time Robust Estimation.- C.1 Introduction.- C.2 Problem Formulation.- C.3 Derivation of the Estimator.- C.4 Related Work.- D Application Data.- D.1 Underwater Vehicle Application.- D.1.1 The Plant.- D.1.2 Description of Robust Filter Design.- D.2 Reentry Vehicle Application.- D.2.1 The Plant.- D.2.2 The Filters.