Orbital Motion Intention Recognition for Space Non-cooperative Targets Based on Incomplete Time Series Data The primary function of this repository is to provide the code and dataset related to the paper "Orbital Motion Intention Recognition for Space Non-cooperative Targets Based on Incomplete Time Series Data." The abstract of the paper is as follows:
"This study focuses on establishing a method for recognizing intentions of non-cooperative targets in orbital dynamics using incomplete time series data. Leveraging the relative orbital dynamics model, the study delineates 38 distinct motion intentions. Subsequently, a neural network-based intention recognition method is developed, conducting a comparative analysis along two explicit dimensions: network architecture and types of time series data. Network architecture encompasses long short-term memory networks (LSTM), gated recurrent unit networks (GRU), and self-attention structures. The types of time series data include position, angular, and distance measurements. Experimental findings demonstrate that combining gated recurrent unit networks with attention mechanisms achieves the highest performance in intention recognition, reaching a maximum accuracy of $ 95\% $. Both position and angular measurement time series exhibit exceptional performance, with recognition accuracies exceeding $ 93\% $, while distance measurement time series are ineffective in intention recognition. Moreover, the study examines the practicality of the proposed method in terms of observation noise and nonlinear dynamics, assessing performance through accuracy and F1 scores. An analysis of intention observability is also conducted to improve model interpretability. This approach holds significant promise for applications such as spacecraft collision avoidance."
The currently provided code and dataset do not fully cover the experiments in the paper. Our team will continue to supplement and update them in the future.
The link to the dataset: https://pan.baidu.com/s/1-NbGr0sTw9K0lZbipuEUqg Extraction code: aom4