1,720,972 research outputs found
Robust and efficient single-CNN-based spacecraft relative pose estimation from monocular images
No full textAutonomous spacecraft relative navigation is crucial for future space missions, but achieving accurate and efficient pose estimation, especially for uncooperative targets, remains challenging. Despite recent strides in AI-based solutions and benchmark datasets, existing algorithms prioritize accuracy over computational efficiency and struggle to generalize from synthetic to real-world scenarios. The paper addresses these issues by proposing a novel pose estimation algorithm from monocular images based on a single multitasking convolutional neural network (CNN) performing both region of interest (ROI) estimation and keypoint regression. The proposed architecture leverages an outlier rejection scheme based on confidence scores to enhance robustness and reliability, while a check on the need for a second inference has been introduced to improve the computational efficiency and potentially enable higher navigation filter update rates. To improve the domain gap bridging capabilities of the introduced pipeline, custom augmentation techniques are presented within this work, including a novel noise augmentation mimicking actual sensor noise. These augmentations have been used during the model training and proved to be beneficial in enhancing the CNN performances, setting the new highest score for object detection and keypoint regression on a benchmark dataset. The performances of the proposed architecture have been assessed using synthetic SPEED images. The outcomes of the analyses demonstrate that our architecture achieves high-level performances on synthetic SPEED images with a mean translational error lower than 10 cm and a mean angular error of about 1.4 degrees, outperforming other more complex models while maintaining an execution time on the CPU of an Apple® SiliconTM M1 Pro processor in the order of 60 ms to 131 ms, depending if one or two inferences are needed to retrieve the pose. Further, the standard deviations of the errors are 11.5 cm and 1.0 degree for the translation and attitude errors, respectively, revealing the high precision of the proposed solution and the absence of strong outliers, especially for the relative attitude estimation where the registered standard deviation is the lowest among the methods available in the literature. The evaluation using mock-up SPEED+ frames confirms the effectiveness of the introduced domain gap reduction strategies, with performance metrics remaining competitive despite increased errors compared to synthetic images mainly due to low illumination conditions. Notably, the paper also outlines the extension of the proposed architecture to more complex targets to prove the adaptability of the proposed approach. The outcomes of this analysis confirm that the introduced pipeline still achieves high accuracy in the relative pose estimation tasks even for targets with complex geometries and a high probability of keypoints occlusions, as for the case of Envisat. Conversely, for highly symmetric targets like VESPA, the performances of the keypoint regression degrades, leading to wrong estimates due to the uncertainties in the retrieved keypoints locations
Tango Spacecraft Wireframe Dataset Model for Line Segments Detection
No full textGeneral Description:
The "Tango Spacecraft Wireframe Dataset Model for Line Segments Detection" dataset here published should be used for line detection and segmentation tasks. It is split into 30002 train images and 3002 test images representing the Tango spacecraft from Prisma mission, being the only publicly available dataset of synthetic space-borne images tailored to line detection tasks (up to our knowledge). The label of each image gives the reprojection of a simplified wireframe model of Tango on the image plane split into lines. The labels are written following the Wireframe Model format. The "Tango Spacecraft Wireframe Dataset Model for Line Segments Detection" is also the largest dataset with wireframe annotations available up to date. More information on the dataset split and on the label format are reported below.
Images Information:
The dataset comprises 30002 synthetic grayscale images of Tango spacecraft from Prisma mission that serves as train set, while the test set is formed by 3002 synthetic grayscale images of Tango spacecraft from Prisma mission in PNG format. About 1/6 of the images both in the train and in the test set have a non-black background, obtained by rendering an Earth-like model in the raytracing process used to define the images reported. The images have different resolutions in pixels and are noise-free to increase the flexibility of the dataset. The illumination direction of the spacecraft in the scene is uniformly distributed in the 3D space in agreement with the Sun position constraints.
Labels Information:
Labels in the Wireframe dataset format are here provided in separated JSON files. The files are formatted per each image as in the following example:
width : 98 # width in pixels (int) of the current image
height : 176 # height in pixels (int) of the current image
lines : [[line1], [line2], ..., [lineN]] # list of lines in each image
filename : tango_img_866.png # string with image name and format
Per each line (line1, ... , lineN) in lines, the format is [x0, y0, x1, y1].
(x0, y0) are the coordinates (float) of the line starting point in the image reference frame (x pointing right and y pointing down with origin located in the top-left corner of the image).
(X1, y1) are the coordinates (float) of the line ending point in the image reference frame (x pointing right and y pointing down with origin located in the top-left corner of the image).
Note that the starting point is assumed to be the left-most endpoint (lower x coordinate in image reference frame) of each line. In the case of vertical lines, the starting point is the upper-most endpoint (lower y coordinate in image reference frame) of each line
Hovering of an Electrically Actuated Spacecraft in a Small-Body Plasma Field
Full text linkThis paper presents simulation models and an analysis of the hovering capability of an electrostatic spacecraft around a small celestial body. The hovering capabilities of an electrically actuated spacecraft are evaluated by combining orbital dynamics analysis with 3-D fully kinetic particle-in-cell simulations of asteroid/spacecraft interactions with the solar-wind plasma. The zero-velocity curves obtained from the analysis allow the identification of the equilibrium points for different levels of charge. The analysis of the system equilibria indicates the presence of equilibrium points in the combined gravitational, electrostatic, and solar illumination fields, most of which can be obtained by charging the spacecraft negatively. The charge-to-mass ratio needed to hover is obtained for different orbital positions, and an analysis of the sensitivity of the equilibria with respect to the spacecraft equivalent radius and with respect to the sun-to-main-body distance provides additional insight into the system dynamics
Robust spacecraft relative pose estimation via CNN-aided line segments detection in monocular images
Full text linkAutonomous spacecraft relative navigation via monocular images became a hot topic in the past few years and, recently, received a further push thanks to the constantly growing field of artificial neural networks and the publication of several spaceborne image datasets. Despite the proliferation of spacecraft relative-state initialization algorithms developed, most architectures adopt computationally expensive solutions relying on convolutional neural networks (CNNs) that provide accurate output at the cost of a high computational burden that seems unfeasible for current spaceborne hardware. The paper addresses this issue by proposing a novel pose initialization algorithm based on lightweight CNNs. Inspired by previous state-of-the-art algorithms, the developed architecture leverages a fast and accurate target detection CNN followed by a line segment detection CNN capable of running with low inference time on mobile devices. The line segments and their junctions are grouped into complex geometrical groups, reducing the solution search space, and subsequently, they are adopted to extract the final pose estimate. As a main outcome, the analyses demonstrate that the lightweight architecture developed scores high accuracy in the pose estimation task, with a mean estimation error of less than 10 cm in translation and 2.5°in rotation. The baseline algorithm scores a mean SLAB error of 0.04552 with a standard deviation of 0.22972 in the test dataset. Detailed analyses demonstrate that the uncertainties on the overall pose score are driven mainly by errors in the relative attitude, which gives the highest contribution to the pose error metric adopted. The analyses on the error distributions point out that the uncertainties on the estimated relative position are higher in the camera boresight axis direction. Concerning the relative attitude, the algorithm proposed has higher uncertainties in estimating directions of the target x and y axes due to ambiguities related to the target geometry. Notably, the target detection CNN trained in this work outperforms the previous top scores in the benchmark dataset. The performances of the proposed algorithm have been investigated further by analyzing the effects on the accuracy due to the relative distance and the presence of background in the images. Lastly, the paper delves into the possibility of adopting a sub-portion of the 2D-to-3D match matrix made by the most complex perceptual groups identified that positively affects the overall run-time, pointing out the performances in terms of accuracy of the estimates and providing a comparison of both the baseline and the reduced match matrix versions against state-of-the-art algorithms concerning relative position and attitude errors and solution availability, highlighting the high accuracy and solution availability of the proposed architectures
Synthetic thermal image generation and processing for close proximity operations
Full text linkThe new scenarios foreseen in forthcoming space missions have increased interest towards optical-based relative navigation techniques, which have demonstrated efficacy in a variety of operational conditions. Although object detection methods have predominantly been used within the visible spectrum, optical payloads struggle in weak lighting conditions and are susceptible to overexposure. Consequently, thermal imaging systems are being investigated as a potential solution, as their integration into the current systems would greatly extend future mission capabilities. This study seeks to fill the gap in literature by assessing the performance of state-of-the-art object detection algorithms with images captured in the thermal spectrum. Given the scarcity of readily available thermal infrared (TIR) images captured in orbit, a novel rendering pipeline is implemented to generate physically accurate thermal images relevant to close-proximity scenarios. These synthetic representations feature a simplified target spacecraft against Earth and deep space backgrounds, including variations in illumination conditions, material properties, relative state, and scale. To ensure realistic outputs, the radiative field of the Earth is modelled based on satellite measurements collected in the cloud and Earth radiant energy system (CERES) database. To enrich the fidelity of the outputs, a thermal sensor model and the corresponding noise levels are introduced in the pipeline. The generated images are then used to test the performance of traditional object detection algorithms in discerning the region of interest (ROI) under different orbital scenarios. The results demonstrate the effectiveness of the selected methodologies in mitigating the influence of the Earth in the ROI extraction process, while also revealing a performance degradation due to the presence of multi-material targets
Closed-loop AI-aided image-based GNC for autonomous inspection of uncooperative space objects
Full text linkAutonomy is increasingly crucial in space missions due to several factors driving the exploration and utilization of space. In the meanwhile, Artificial Intelligence methods begin to play a crucial role in addressing the challenges associated with and enhancing autonomy in space missions. The proposed work develops a closed-loop simulator for proximity operations scenarios, particularly for the inspection of an unknown and uncooperative target object, with a fully AI-based image processing and GNC chain. This tool is based on four main blocks: image generation, image processing, navigation filter, and guidance and control blocks. All of them have been separately tested and tuned to ensure the correct interface and compatibility in the close-loop architecture. Afterwards, the overall architecture is deployed in an extensive Montecarlo testing campaign to verify and validate the performance of the proposed IP-GNC loop
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
- …
