Optimal Estimation of Dynamic Systems

Optimal Estimation of Dynamic Systems, Second Edition highlights the importance of both physical and numerical modeling in solving dynamics-based estimation problems found in engineering systems. Accessible to engineering students, applied mathematicians, and practicing engineers, the text presents the central concepts and methods of optimal estimation theory and applies the methods to problems with varying degrees of analytical and numerical difficulty. Different approaches are often compared to show their absolute and relative utility. The authors also offer prototype algorithms to stimulate the development and proper use of efficient computer programs. MATLAB® codes for the examples are available on the book’s website.
New to the Second EditionWith more than 100 pages of new material, this reorganized edition expands upon the best-selling original to include comprehensive developments and updates. It incorporates new theoretical results, an entirely new chapter on advanced sequential state estimation, and additional examples and exercises.

An ideal self-study guide for practicing engineers as well as senior undergraduate and beginning graduate students, the book introduces the fundamentals of estimation and helps newcomers to understand the relationships between the estimation and modeling of dynamical systems. It also illustrates the application of the theory to real-world situations, such as spacecraft attitude determination, GPS navigation, orbit determination, and aircraft tracking.

John L. Crassidis, Ph.D., is a professor of mechanical and aerospace engineering and the associate director of the Center for Multisource Information Fusion at the University at Buffalo, State University of New York. He previously worked at Texas A&M University, the Catholic University of America, and NASA’s Goddard Space Flight Center, where he contributed to attitude determination and control schemes for numerous spacecraft missions.
John L. Junkins, Ph.D., is a distinguished professor of aerospace engineering and the founder and director of the Center for Mechanics and Control at Texas A&M University. In addition to his historical contributions in analytical dynamics and spacecraft GNC, Dr. Junkins and his team have designed, developed, and demonstrated several new electro-optical sensing technologies.

Least Squares ApproximationA Curve Fitting ExampleLinear Batch EstimationLinear Sequential EstimationNonlinear Least Squares EstimationBasis FunctionsAdvanced Topics

Probability Concepts in Least SquaresMinimum Variance EstimationUnbiased EstimatesMaximum Likelihood EstimationCramer-Rao InequalityConstrained Least Squares Covariance Maximum Likelihood EstimationProperties of Maximum Likelihood EstimationBayesian EstimationAdvanced Topics

Sequential State EstimationA Simple First-Order Filter ExampleFull-Order EstimatorsThe Discrete-Time Kalman FilterThe Continuous-Time Kalman FilterThe Continuous-Discrete Kalman FilterExtended Kalman FilterUnscented FilteringConstrained Filtering

Advanced Topics in Sequential State EstimationFactorization MethodsColored-Noise Kalman FilteringConsistency of the Kalman FilterConsider Kalman FilteringDecentralized FilteringAdaptive FilteringEnsemble Kalman FilteringNonlinear Stochastic Filtering TheoryGaussian Sum FilteringParticle FilteringError AnalysisRobust Filtering

Batch State EstimationFixed-Interval SmoothingFixed-Point SmoothingFixed-Lag SmoothingAdvanced Topics

Parameter Estimation: ApplicationsAttitude DeterminationGlobal Positioning System NavigationSimultaneous Localization and MappingOrbit DeterminationAircraft Parameter IdentificationEigensystem Realization Algorithm

Estimation of Dynamic Systems: ApplicationsAttitude EstimationInertial Navigation with GPSOrbit EstimationTarget Tracking of AircraftSmoothing with the Eigensystem Realization Algorithm

Optimal Control and Estimation TheoryCalculus of VariationsOptimization with Differential Equation ConstraintsPontryagin’s Optimal Control Necessary ConditionsDiscrete-Time ControlLinear Regulator ProblemsLinear Quadratic-Gaussian ControllersLoop Transfer RecoverySpacecraft Control Design

Appendix A: Review of Dynamical SystemsAppendix B: Matrix PropertiesAppendix C: Basic Probability ConceptsAppendix D: Parameter Optimization MethodsAppendix E: Computer Software

Index

A Summary appears at the end of each chapter.