Orbital Error Analysis and Correction Method based on Data Mining for Multiple Kinetic Energy Impactors

Abstract

In asteroid defense, multiple launches of kinetic energy impactors require collaborative control to improve the deflection efficiency. Therefore, an orbital error analysis and correction method of multi kinetic energy impactors based on data mining is proposed. Firstly, the operating characteristics are extracted from the operating data of impactors. The random forest method in data mining is used to mine the error to judge whether there is an error in the operating data. Then, according to the analysis error, the orbital error correction is implemented by using the correction method based on the controller to return the impactors to the expected orbit. Finally, the simulation results show that the AUC value under the ROC curve of this method is 0.9225, which is closer to 1.0 compared with other methods, indicating that the accuracy of error analysis is higher. After the correction of this method, the flight orbit is more consistent with the expected orbit.