Multi-model self-adaptive state estimation method and system based on ls

Abstract

A multi-model adaptive state estimation method based on LS, including: S1. Acquiring the observation sequence of each sensor on the same moving target, and forming the observation sequence of the moving target in the X, Y, and Z directions in the space Cartesian coordinate system at different times ; S2, determine the number and number of polynomial equations with time as a variable, and estimate the polynomial coefficients of the least squares LS to the observation sequence of the moving target in the X, Y, and Z directions respectively; S3, according to the polynomial equation in step S2, Calculate the regression sum of squares and the residual sum of squares respectively and perform F test to determine the best fitting polynomial; S4, determine the corrected best fitting polynomial; S5, calculate the first partial derivative and the second partial derivative for time respectively, and obtain the moving target Corresponding velocity and acceleration polynomials; S6. Carry out coordinate conversion on the target position state estimate, obtain the corresponding position coordinates, and then combine the velocity estimate value to complete the target state estimation.

Description

technical field [0001] The invention relates to the technical field of data fusion, in particular to an LS-based multi-model self-adaptive state estimation method and system. Background technique [0002] Multi-sensor data fusion is an automatic information comprehensive processing technology formed and developed in the 1980s. It has developed into a very active and popular research field in national defense, and it is the common concern of multi-disciplinary, multi-department and multi-field High-level common key technology, many countries have listed it as the key technology for the next stage of key development. [0003] One of the main tasks of multi-sensor data fusion is to use multi-sensor data to estimate the state of the target. State estimation is to find the state vector that best fits the observed data through mathematical methods. State estimation mainly refers to position and speed estimation, which includes state estimation of moving targets based on single-s...

AI Technical Summary

Problems solved by technology

[0004] The state estimation of the maneuvering target belongs to the problem of nonlinear filtering estimation. The basic idea of ​​traditional nonlinear filtering is to approximate the nonlinearity of the system in a parametric analytical form, which inevitably introduces linearization errors. When the error increases to a certain To a certain extent, the accuracy of state estimation will be reduced, and in severe cases, it will lead to problems such as filter divergence

Images