An Efficient Pose Estimation Algorithm for Non-Cooperative Space Objects Based on Dual-Channel Transformer
Non-cooperative space object pose estimation is a key technique for spatial on-orbit servicing, where pose estimation algorithms based on low-quality, low-power monocular sensors provide a practical solution for spaceborne applications. The current pose estimation methods for non-cooperative space o...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-11-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/22/5278 |
| _version_ | 1827638862220886016 |
|---|---|
| author | Ruida Ye Yuan Ren Xiangyang Zhu Yujing Wang Mingyue Liu Lifen Wang |
| author_facet | Ruida Ye Yuan Ren Xiangyang Zhu Yujing Wang Mingyue Liu Lifen Wang |
| author_sort | Ruida Ye |
| collection | DOAJ |
| description | Non-cooperative space object pose estimation is a key technique for spatial on-orbit servicing, where pose estimation algorithms based on low-quality, low-power monocular sensors provide a practical solution for spaceborne applications. The current pose estimation methods for non-cooperative space objects using monocular vision generally consist of three stages: object detection, landmark regression, and perspective-n-point (PnP) solver. However, there are drawbacks, such as low detection efficiency and the need for prior knowledge. To solve the above problems, an end-to-end non-cooperative space object pose estimation learning algorithm based on dual-channel transformer is proposed, a feature extraction backbone network based on EfficientNet is established, and two pose estimation subnetworks based on transformer are also established. A quaternion SoftMax-like activation function is designed to improve the precision of orientation error estimating. The method only uses RGB images, eliminating the need for a CAD model of the satellite, and simplifying the detection process by using an end-to-end network to directly detect satellite pose information. Experiments are carried out on the SPEED dataset provided by the European Space Agency (ESA). The results show that the proposed algorithm can successfully predict the satellite pose information and effectively decouple the spatial translation information and orientation information, which significantly improves the recognition efficiency compared with other methods. |
| first_indexed | 2024-03-09T16:29:44Z |
| format | Article |
| id | doaj.art-bed2fa9dd8914feab7eee9b58ab1af6c |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T16:29:44Z |
| publishDate | 2023-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-bed2fa9dd8914feab7eee9b58ab1af6c2023-11-24T15:04:09ZengMDPI AGRemote Sensing2072-42922023-11-011522527810.3390/rs15225278An Efficient Pose Estimation Algorithm for Non-Cooperative Space Objects Based on Dual-Channel TransformerRuida Ye0Yuan Ren1Xiangyang Zhu2Yujing Wang3Mingyue Liu4Lifen Wang5Department of Aerospace Engineering and Technology, Space Engineering University, Beijing 101416, ChinaDepartment of Basic Course, Space Engineering University, Beijing 101416, ChinaDepartment of Aerospace Engineering and Technology, Space Engineering University, Beijing 101416, ChinaDepartment of Basic Course, Space Engineering University, Beijing 101416, ChinaDepartment of Aerospace Engineering and Technology, Space Engineering University, Beijing 101416, ChinaDepartment of Aerospace Engineering and Technology, Space Engineering University, Beijing 101416, ChinaNon-cooperative space object pose estimation is a key technique for spatial on-orbit servicing, where pose estimation algorithms based on low-quality, low-power monocular sensors provide a practical solution for spaceborne applications. The current pose estimation methods for non-cooperative space objects using monocular vision generally consist of three stages: object detection, landmark regression, and perspective-n-point (PnP) solver. However, there are drawbacks, such as low detection efficiency and the need for prior knowledge. To solve the above problems, an end-to-end non-cooperative space object pose estimation learning algorithm based on dual-channel transformer is proposed, a feature extraction backbone network based on EfficientNet is established, and two pose estimation subnetworks based on transformer are also established. A quaternion SoftMax-like activation function is designed to improve the precision of orientation error estimating. The method only uses RGB images, eliminating the need for a CAD model of the satellite, and simplifying the detection process by using an end-to-end network to directly detect satellite pose information. Experiments are carried out on the SPEED dataset provided by the European Space Agency (ESA). The results show that the proposed algorithm can successfully predict the satellite pose information and effectively decouple the spatial translation information and orientation information, which significantly improves the recognition efficiency compared with other methods.https://www.mdpi.com/2072-4292/15/22/5278non-cooperative space objectspose estimationmonocular sensordual-channel transformerquaternion SoftMax-like activation function |
| spellingShingle | Ruida Ye Yuan Ren Xiangyang Zhu Yujing Wang Mingyue Liu Lifen Wang An Efficient Pose Estimation Algorithm for Non-Cooperative Space Objects Based on Dual-Channel Transformer Remote Sensing non-cooperative space objects pose estimation monocular sensor dual-channel transformer quaternion SoftMax-like activation function |
| title | An Efficient Pose Estimation Algorithm for Non-Cooperative Space Objects Based on Dual-Channel Transformer |
| title_full | An Efficient Pose Estimation Algorithm for Non-Cooperative Space Objects Based on Dual-Channel Transformer |
| title_fullStr | An Efficient Pose Estimation Algorithm for Non-Cooperative Space Objects Based on Dual-Channel Transformer |
| title_full_unstemmed | An Efficient Pose Estimation Algorithm for Non-Cooperative Space Objects Based on Dual-Channel Transformer |
| title_short | An Efficient Pose Estimation Algorithm for Non-Cooperative Space Objects Based on Dual-Channel Transformer |
| title_sort | efficient pose estimation algorithm for non cooperative space objects based on dual channel transformer |
| topic | non-cooperative space objects pose estimation monocular sensor dual-channel transformer quaternion SoftMax-like activation function |
| url | https://www.mdpi.com/2072-4292/15/22/5278 |
| work_keys_str_mv | AT ruidaye anefficientposeestimationalgorithmfornoncooperativespaceobjectsbasedondualchanneltransformer AT yuanren anefficientposeestimationalgorithmfornoncooperativespaceobjectsbasedondualchanneltransformer AT xiangyangzhu anefficientposeestimationalgorithmfornoncooperativespaceobjectsbasedondualchanneltransformer AT yujingwang anefficientposeestimationalgorithmfornoncooperativespaceobjectsbasedondualchanneltransformer AT mingyueliu anefficientposeestimationalgorithmfornoncooperativespaceobjectsbasedondualchanneltransformer AT lifenwang anefficientposeestimationalgorithmfornoncooperativespaceobjectsbasedondualchanneltransformer AT ruidaye efficientposeestimationalgorithmfornoncooperativespaceobjectsbasedondualchanneltransformer AT yuanren efficientposeestimationalgorithmfornoncooperativespaceobjectsbasedondualchanneltransformer AT xiangyangzhu efficientposeestimationalgorithmfornoncooperativespaceobjectsbasedondualchanneltransformer AT yujingwang efficientposeestimationalgorithmfornoncooperativespaceobjectsbasedondualchanneltransformer AT mingyueliu efficientposeestimationalgorithmfornoncooperativespaceobjectsbasedondualchanneltransformer AT lifenwang efficientposeestimationalgorithmfornoncooperativespaceobjectsbasedondualchanneltransformer |