Star tracker algorithm improvements to restore performance after radiation exposure
Bibliographic record
Abstract
Rocket Lab’s ST16-RT2 star tracker uses a CMOS detector to image stars in the sensor’s field of view, software to isolate the stars in the image from the background, and an on-board catalogue to identify the stars in the image and, thus, determine the attitude of the satellite. High temperatures and radiation effects raise the black level and standard deviation of the noise on the sensor. This results in a loss of dynamic range and complicates the pixel classification between star candidates and the background. Pixels improperly designated as part of a star are referred to as false positives, and pixels improperly designated as the background are false negatives. A high quantity of false positives adds to the processing time required to identify stars in an image and has the potential to reduce the accuracy of the attitude solutions provided by the star tracker. To maximize the dynamic range available to a specific sensor, Rocket Lab calibrates analog offsets to the device at room temperature. In higher temperature and radiation environments, the star tracker’s dynamic range is reduced and background noise increases. This increases the quantity of false-positive pixels which, in turn, limits star tracker performance. Utilizing analog offsets determined on-chip and a more robust thresholding algorithm, the star tracker can recover dynamic range and more effectively suppress noise. Such improvements would extend a star tracker's life or allow the product to be used in more extreme environments, without compromising the hardware heritage. To implement these improvements, the limitations of the current thresholding algorithm were analyzed, and a new algorithm was developed. Dark image data was collected using non-irradiated and irradiated detectors, with both the preset and on-chip analog offsets, at temperatures ranging from 20 °C to 70 °C. Processing the dark images using this new approach showed a significant reduction in false-positive pixels recorded using the irradiated boards compared to the current algorithm. The reduction in false positives prevented the lit pixel buffer from filling until 64°C, a 30 °C increase in performance for the irradiated boards. Images were then collected using simulated star fields at various slew rates to determine if the reduction in false positives led to an increase in false negatives. Thresholding parameters were tuned and recommended configurations were given for on-orbit testing.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.008 | 0.006 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".