Estimators for uncertain dynamic systems
Generally speaking, a dynamic system with uncertainty means that there are perturbations, deterministic or stochastic, which are not exactly specified, and there are unknown characteristics which have to be estimated. The book is devoted to various optimal estimation problems for such dynamic systems which are assumed to be linear. Even in this case the problems are quite complex and are reduced to different variational problems. The solution to the latter depends on the specific model process, the type of uncertainty and the estimation performance criteria. Hence the optimal estimator might come as the solution to a variational problem. Not an easy task, in general. When stydying dynamic systems under uncertainty, the author follows a very natural idea: to solve the so-called guaranteed estimation problem and do this in an easy way, e.g. by using simplified algorithms. The goal however is that these estimators, not being exactly optimal, have to obey at least some guaranteed ``good properties. The next important question is how to find the level of nonoptimality when using simplified algorithms and at the same time avoid, if possible, the original complicated variational problems. The material is distributed into five chapters. The models in Chapter 1 are deterministic, the problems are parametric and the arguments, when finding the guaranteed estimators, are based on linear programming, least squares method or appropriate functional approximations. Chapter 2 deals with deterministic dynamic systems with uncertainty and their guaranteed estimators. Lagrange's principle, duality and other classical results are among the used tools. Chapters 3 and 4 deal with stochastic systems perturbed by a noise like the white Gaussian noise, so here we see results in the spirit of the Kalman filters and their modifications. Finally, Chapter 5 treats dynamic systems with aftereffects, e.g. systems with delay, and several estimation problems for them. Perhaps these are among the most difficult problems so we have to go to simplified algorithms and look for guaranteed estimators. At the end of each chapter there are short sections, one of which contains useful comments about the progress in the area and the other one contains exercises. The book ends with an extensive references list and index. Thus, a large number of different dynamic models, deterministic and stochastic, are treated in detail. For each of the models the author has exhibited a solution to the guaranteed estimation problem deriving also useful properties of the guaranteed estimators. All cases are well illustrated with examples coming from the real practice. Related topics such as computational aspects are also discussed. Obviously, the reader needs a good knowledge in some areas of mathematics (functional analysis, probability, elements of stochastic processes) in order to understand the main results and ideas and in perspective to apply them when treating other theoretical and practical problems. The book is addressed to applied mathematicians and eventually can be used when teaching a special course to postgraduate students.
- On one approach to the solution of the problem of guaranteeing estimation for Volterra equations
- Determination of worst signals in problems of guaranteed estimation
- Comparison of stochastic and guaranteed approaches to the estimation of the state of dynamic systems
- Optimizing estimation of a statistically undefined system
- Synthesis of state unknown inputs observers for nonlinear Lipschitz systems with uncertain disturbances
- A posteriori minimax estimation with likelihood constraints
- Minimax filtering in a stochastic differential system with non-stationary perturbations of unknown intensity
- Levels of nonoptimality of the Weiszfeld algorithm in the least-modules method
- Minimax-statistical approach to increasing reliability of measurement information processing
- Synthesis of reduced Kalman filter with the guaranteed estimation quality of dynamic system state
- Calibration of an accelerometer unit with asymmetric models of readings of sensors
- Nonsmooth variational problems for calibration of accelerometer units
- RAP-method (random perturbation method) for finding \(S\)-minimax control vectors and parameter estimates for some linear systems with random coefficients
- Minimax estimation by probabilistic criterion
- Comparison of linear and nonlinear methods of confidence estimation for statistically uncertain systems
- Filtration of a random process in a statistically uncertain linear stochastic differential system
- State observer synthesis by measurement results for nonlinear Lipschitz systems with uncertain disturbances
- Uncertainty quantification in hybrid dynamical systems
- Minimax filtering in linear stochastic uncertain discrete-continuous systems
- Guaranteed approach for determining the optimal design of accelerometer unit calibration
- scientific article; zbMATH DE number 6692868 (Why is no real title available?)
- scientific article; zbMATH DE number 3850958 (Why is no real title available?)
- A statistical minimax approach to optimizing linear models under a priori uncertainty conditions
- scientific article; zbMATH DE number 1203306 (Why is no real title available?)
- The Kalman-Bucy filter accuracy in the guaranteed parameter estimation problem with uncertain statistics
- scientific article; zbMATH DE number 1748552 (Why is no real title available?)
- scientific article; zbMATH DE number 926780 (Why is no real title available?)
- scientific article; zbMATH DE number 227722 (Why is no real title available?)
- Filtering Problem for Discrete Volterra Equations with Combined Disturbances
- On a Boundary Value Problem in the Filtering Theory for Discrete Volterra Equations
- Optimization approach to guaranteed state estimation of dynamic systems
- Estimation of systems with statistically-constrained inputs
- Tight computationally efficient approximation of matrix norms with applications
- Minimizing sequences for a family of functional optimal estimation problems
- Estimation of continuously differentiable signal with allowance for constraints
- Constructing maximum likelihood estimates for statistically uncertain linear systems
- Robust signal recovery under uncertain-but-bounded perturbations in observation matrix
- Guaranteed state estimation for with mixed-uncertainty linear dynamical systems.
- An iterative algorithm for \(l_1\)-norm approximation in dynamic estimation problems
- Minimax linear estimation with the probability criterion under unimodal noise and bounded parameters
- Minimax estimation for singular linear multivariate models with mixed uncertainty
- Guaranteed parameter estimation in unstable dynamic systems
- Ellipsoid-based parametric estimation in the linear multidimensional systems with uncertain model description
- Properties of nonlinear confidence estimates for statistically uncertain systems
- Minimax nature of the linear estimates of the indefinite stochastic vector from the generalized probabilistic criteria
This page was built for publication: Estimators for uncertain dynamic systems
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4486011)
