Algorithms for Adaptive Evolution of Stochastic Control Objects

The expansion and constant complication of the scope of research objects requires the development of new effective methods and algorithms for adaptive identification, state assessment and management in uncertainty conditions based on the concept of conditional Gaussian filtering. Also, in the non-parametric description of correlation interactions requires the development of systematic algorithms for conditional-optimal filtering of control objects and adaptive assessment of the state of control objects, taking into account parametric upheavals. In addition, it would be expedient to develop stable algorithms for suboptimal assessment of the state of control objects, extended state vectors of controlled objects, as well as stable algorithms for multi-step assessment of the state of nonlinear control objects. The monograph considers the development of algorithms and computational schemes for adaptive assessment of the state of stochastic control objects on the basis of the conditional Gaussian filtering method and methods of their practical application.