Efficient Star Identification Using a Neural Network

Abstract

<jats:p>The required precision for attitude determination in spacecraft is increasing, providing a need for more accurate attitude determination sensors. The star sensor or star tracker provides unmatched arc-second precision and with the rise of micro satellites these sensors are becoming smaller, faster and more efficient. The most critical component in the star sensor system is the lost-in-space star identification algorithm which identifies stars in a scene without a priori attitude information. In this paper, we present an efficient lost-in-space star identification algorithm using a neural network and a robust and novel feature extraction method. Since a neural network implicitly stores the patterns associated with a guide star, a database lookup is eliminated from the matching process. The search time is therefore not influenced by the number of patterns stored in the network, making it constant (O(1)). This search time is unrivalled by other star identification algorithms. The presented algorithm provides excellent performance in a simple and lightweight design, making neural networks the preferred choice for star identification algorithms.</jats:p>

An Evaluation of Low-Cost Vision Processors for Efficient Star Identification

“…Moreover, time expensive database searches required in the case of classical star identification algorithms are removed from the process since the star patterns are implicitly saved in the network. The algorithm also proves to be resilient to noise and false stars [6].…”

“…There are two of these main categories-pattern and subgraph isomorphism-based feature extraction [5]. In this manuscript, a pattern feature extraction based method as presented in [6], is used. The pattern feature extraction algorithms assign each reference star to a pattern based on the stars surrounding it and find the closest matching pattern in a database.…”

“…A simple pattern feature extraction method is used in this work as proposed in [6]. Specifically, a rotational-invariant pattern feature extraction is implemented since a neural network requires an input of invariant size and order.…”

“…• The two processors-the Myriad 2 VPU and the STM32 Microcontroller have been chosen to perform evaluations concerning the accuracy, frame rate, and power. This helps in understanding which processor will be best suited for the low-power consumption star tracker • An efficient and robust NN-based star identification algorithm [6] has been used in this work to perform the testing with the two processors. This algorithm has high analytical performance and is robust to false stars.…”

“…Recently, a new lost in space star identification algorithm similar to [7] has been proposed in [6] that performs the star tracking task using an NN Artificial Intelligence (AI) based approach. This approach is capable of achieving and exceeding the performance of traditional algorithmic-based approaches on synthetic datasets.…”