In the realm of energy and space exploration, a team of researchers from institutions including the National Astronomical Observatory of Japan, the University of Tokyo, and the University of Hawaii have developed a novel approach to detect moving objects in the solar system. This advancement, published in the journal Astronomy & Computing, could significantly reduce the need for manual verification, a process that has been both time-consuming and costly.
The researchers have proposed a multi-input convolutional neural network (CNN) integrated with a convolutional block attention module. This method is designed to enhance the detection of moving objects, such as asteroids or space debris, from wide-field survey data. The new model introduces two key innovations. First, it processes multiple stacked images simultaneously, allowing for more efficient learning from multiple inputs. Second, it incorporates a convolutional block attention module, which enables the model to focus on essential features in both spatial and channel dimensions.
The performance of this model was evaluated on a dataset consisting of approximately 2,000 observational images. The results were impressive, with an accuracy of nearly 99% and an Area Under the Curve (AUC) of greater than 0.99. These metrics indicate that the proposed model achieves excellent classification performance. By adjusting the threshold for object detection, the new model reduces the human workload by more than 99% compared to manual verification.
For the energy sector, particularly in space-based solar power or satellite management, this technology could be a game-changer. The ability to accurately and efficiently detect moving objects can enhance the safety and effectiveness of space operations. It can also aid in the monitoring and management of space debris, which is a growing concern for the energy industry as it relies more on space-based assets. This research demonstrates the potential of advanced machine learning techniques to address real-world challenges in the energy and space sectors.
This article is based on research available at arXiv.