基于光学与激光观测的空间碎片旋转状态反演方法研究

In order to curb the future growth of space debris, the active debris removal has become one of the necessary measures. Before the implementation of active removal tasks, rotational states of the space debris need to be obtained on the ground in advance for the development of the strategy. At present, the estimation methods of the rotational states of the space debris can be divided into two types: shape-independent method and shape-dependent method. Based on the optical and laser observation, the research in the project will carry out the analyses of the observational data to obtain the rotational states of the space debris for different types of targets. On the research of the shape-independent estimation method, the laser ranging data of the specific space debris with corner cube reflectors will be analyzed for the frequency-domain and additional information. On the research of the shape-dependent estimation method, the intelligent optimization algorithms, such as neural network and genetic algorithm, are introduced to establish the correlations of the rotational states and the observed data, and to achieve the photometric data classification and optimal fitting. The results of the research can give the observational strategy for the rotational state estimation and the applied pattern of the numerical simulation or laboratory model data, and it is expected to make a breakthrough in the rotational state estimation method of the space debris.

为了遏制未来空间碎片的增长，空间碎片的主动清除已成为空间可持续利用的必要措施之一。在执行主动清除任务时，需要预先在地面获取空间碎片的旋转和姿态信息来协助清除策略的制定。目前空间碎片旋转状态反演方法大致可分为两种类型：形状独立型方法与形状依赖型方法。本项目将基于光学与激光观测数据模式，针对不同类型的空间碎片目标，开展利用观测数据获取其旋转状态方法的研究。通过选取特定的带有激光角反射器的空间碎片进行激光测距观测，并对激光测距数据进行频域分析以及额外信息的挖掘，开展形状独立型旋转状态反演方法的研究；通过引入神经网络、遗传算法等智能优化算法，建立旋转状态与观测数据之间的相关性，对光度数据进行分类和最优化拟合匹配，从而实现形状依赖型旋转状态反演。本项目研究结果可以给出反演所需观测数据的采集策略，以及数值模拟或实验室仿真数据的组合利用方式，并有望在空间碎片旋转状态反演技术方面实现突破。

本项目针对项目计划书拟定的研究内容逐项开展研究，在空间碎片旋转状态反演技术方面取得一定进展。在形状独立型方法中，利用天宫一号的激光观测数据成功反演得到了其再入前最后5个月的旋转状态及其演化规律，为极低轨区域的力矩模型研究提供了重要支撑；在形状依赖型方法中，基于Envisat光度观测数据拟合获取其转速变化，并反演得到其转轴分布情况。项目研究结果可以给出反演所需观测数据的采集策略，以及数值模拟或模型仿真数据的组合利用方式。