| 其他摘要 | With the rapid expansion of the number of space objects due to the development of various countries’ large constellation plans, the risk of explosion or collision and breakup events is also increasing sharply. Therefore, vigorously developing new technologies for tracking and monitoring space debris can effectively enhance the collision warning capability for space debris and ensure the safe and smooth development of space activities. Currently, debris laser ranging is booming, with ranging accuracy reaching the decimeter level. Its high-precision laser ranging data can provide higher precision debris orbit information. The main challenges of space debris laser ranging include low prediction accuracy and weak laser echo signals due to diffuse reflection, thus requiring high requirements for pointing and signal-to-noise ratio. The large prediction uncertainty is reflected in two aspects: a large range uncertainty may result in the single-photon detector in the receiving system not detecting the echo photons during the open time; and a large visual position uncertainty may cause the emitted laser to not cover the target.In response to the initial observation and search problem of debris laser ranging, this paper, based on the evolution of orbital uncertainty, mainly completes the following work: Firstly, a method combining Unscented Transform (UT) with Gaussian Mixture Model (GMM) is proposed for orbital uncertainty propagation. Assuming that the initial uncertainty follows a Gaussian distribution, after nonlinear propagation through the dynamic system, the uncertainty will exhibit non-Gaussian characteristics. Therefore, the GMM is used to approximate the initial probability density function, and the UT method is used to forecast sub-Gaussians. Finally, the sub-Gaussians are merged to obtain the accurate terminal probability density function. By comparing with Monte Carlo method results, the accuracy of this method is verified (prediction range error not exceeding 20m, angle error within 2 arcseconds), and while ensuring accuracy, it effectively reduces the calculation time. The calculation time of this method is within 1 minute, while Monte Carlo simulation requires more than two hours, improving the calculation efficiency by more than a hundred times. Secondly, based on the obtained uncertainty results, the concept of capture probability is further proposed and calculated. This concept helps to assess the likelihood of the target being observed during the search process, providing an important reference for subsequent path planning and decision-making. Finally, a search path based on the uncertainty results is proposed. By calculating the capture probability and comparing it with the search path commonly corrected by time uncertainty, a comprehensive evaluation of the proposed target search path is IIIResearch on Space Debris Searching Based on Nonlinear Uncertainty Propagation conducted, indicating that the new search path is more reasonable. The work in this paper can also be applied to the search for laser ranging targets in areas where optical guidance is not possible due to the Earth’s shadow. |
修改评论