Synthetic photometric landmarks used for absolute navigation near an asteroid

Abstract

The need for autonomous location estimation in the form of optical navigation is an essential requirement for forthcoming deep space missions. While crater-based navigation might work well with larger bodies littered with craters, small sub-kilometer bodies do not necessarily have them. We have developed a new pose estimation method for absolute navigation based on photometric local feature extraction techniques thus making it suitable for missions that cannot rely on craters. The algorithm can be used by a navigation filter in conjunction with relative pose estimation such as visual odometry for additional robustness and accuracy. To estimate the position and orientation of the spacecraft in the asteroid-fixed coordinate frame, it uses navigation camera images in combination with other readily available information, such as orientation relative to the stars and the current time for an initial estimate of the asteroid rotation state. Evaluation of the algorithm when using different feature extractors is performed, on one hand, using Monte Carlo simulations and, on the other hand, using actual images taken by the Rosetta spacecraft orbiting the comet 67P/Churyumov–Gerasimenko. Our analysis, where four different feature extraction methods (AKAZE, ORB, SIFT, SURF) were compared, showed that AKAZE is most promising in terms of stability and accuracy.

CNN-Based Local Features for Navigation Near an Asteroid

“…Due to the challenges of creating maplets on board and the limited performance of traditional photometric features, different approaches have been suggested. One such approach is Synthetic Photometric Landmarks (SPLs) [5], which involves rendering a global shape model using a priori information. Traditional AKAZE features are then extracted and matched between the query image and the synthetic image.…”

“…The images acquired during these missions served as the basis for training our feature extractor. Additionally, we supplemented our dataset with synthetic data generated using a Bennu shape model [69] and OpenGL-based rendering software [5], [70]. Images from various missions can be accessed through the NASA Planetary Data System (PDS) or, in the case of the Rosetta mission, via the ESA Planetary Science Archive [71].…”

“…To address concerns about lighting robustness, we incorporate synthetic image pairs that exhibit significant variations in lighting. These synthetic images are generated using an existing OpenGL-based image rendering pipeline, as introduced by Knuuttila et al [5], and integrated with SISPO [70].…”

“…Feature extraction plays a crucial role in most vision-based navigation methods in the field of robotics, including simultaneous localization and mapping (SLAM), as well as absolute navigation approaches like on-board rendering-based Synthetic Photometric Landmarks (SPLs) [5]. Additionally, SLAM methods can incorporate pre-built feature databases.…”

“…Our code together with the trained models are available at https://github.com/oknuutti/navex, while the data used [6] is published through Zenodo.…”