Spacecraft attitude stability judgement method utilizing RCS (radar cross section)

Abstract

The invention relates to a spacecraft attitude stability judgement method utilizing RCS. The method is characterized in that the steps are as follows: when the format of a RCS sequence is correct, the RCS sequence is segmented, discrete wavelet transform is then carried out, so that a discrete wavelet transform coefficient is obtained, and wavelet transform characteristics are extracted; the transform characteristics are normalized and then inputted into a three-layer BP (back-propagation) neural network, and when a BP neural network error function E (W) reaches a minimum value, the attitude stability of a spacecraft is judged according to the output value of the BP neural network; 0 in the output value represents attitude stabilization, and 1 in the output value represents attitude rolling. The method can utilize the characteristics to judge the attitude stability of the spacecraft, and measured data prove that the space target attitude stability identification rate of the method is more than 90 percent and that the operation speed is high ( single target identification time is less than 2s).

Description

technical field [0001] The invention belongs to the field of aerospace measurement and control, and relates to a method for judging the attitude stability of a spacecraft by using RCS, which is suitable for judging the attitude stability of a spacecraft by using radar RCS measurement time series. Background technique [0002] Ground-based radar detection is one of the main means of non-cooperative space target surveillance. During the failure of the spacecraft, especially when the downlink telemetry of the satellite is interrupted, the aerospace measurement and control network can no longer complete the analysis and judgment of the satellite failure, and it is necessary to measure the target through non-cooperative measurement means, which is different from the non-cooperative measurement in the normal state of the spacecraft. Data comparison, analysis and judgment of the failure target state, provide technical support for the decision-making of the spacecraft failure rescue...

AI Technical Summary

Problems solved by technology

Using high-order statistics to identify space targets by Ma Junguo et al. , using multiple hypothesis testing as a classifier to identify three types of typical space objects: three-axis stabilized satellites, spin-stabilized satellites and debris, but this method is based on simulation data, and the measured data proves that its recognition rate is low

Xu Xin et al. used wavelet transform to extract features such as RCS mean value, effective rank, and maximum singular value of space targets, and used fuzzy classification to identify three-axis stable and spin-stabilized targets. When this method selects experimental data from single-turn tracking data for identification, The distance, angle and other information between the space target and the radar are relatively stable, and the category information carried in the data is relatively small, resulting in a low recognition rate (less than 60%) and low operating efficiency of fuzzy recognition

Images