Квазиоптимальные законы управления на основе редукции экстремальных задач в динамических алгоритмах оценки положения маневрирующих объектов

Abstract

A method is proposed for dynamically estimating the position of maneuvering objects based on the use of quasi-optimal control laws in the decomposition mode to construct a model of a dynamic system that describes the movement of maneuvering objects, approximating the real trajectory. The expansion of the state space using the shaping filter method ensures that the regular component of the random maneuvering process is taken into account. Based on the obtained stochastic Markov motion model, a Kalman-type algorithm is proposed for dynamically estimating the position of a maneuvering object using a finite-dimensional approximation. The expressions for the transition matrices of the state and perturbation necessary for the construction of the algorithm are determined. An analysis of the results of numerical simulation suggests an increase in the accuracy of estimating the position of a maneuvering object when using a dynamic estimation algorithm based on quasi-optimal control laws in comparison with a filter based on the Singer model.