Abstract
In this work computer vision algorithms are applied to process BGUSAT SWIR images and generate single-frame attitude quaternions. Two types of image processing software were implemented for image registration. ENVI has the drawback of requiring a manual selection of seed tie points. The MATLAB subroutines are automated and lend themselves to autonomous on-board vision-based single-frame attitude estimation. The proposed methodology was successfully tested on a batch of several images taken at various epoch times. Results analysis of the image processing and of the attitude estimation shows that good performances are achieved, and that the method is robust to variations in the base images. The proposed vision-based attitude estimator is expected to yield better accuracy than estimates based on Sun and magnetic field measurements, by at least one order of magnitude.
Original language | English |
---|---|
State | Published - 1 Jan 2023 |
Event | 62nd Israel Annual Conference on Aerospace Sciences, IACAS 2023 - Haifa, Israel Duration: 15 Mar 2023 → 16 Mar 2023 |
Conference
Conference | 62nd Israel Annual Conference on Aerospace Sciences, IACAS 2023 |
---|---|
Country/Territory | Israel |
City | Haifa |
Period | 15/03/23 → 16/03/23 |
All Science Journal Classification (ASJC) codes
- Aerospace Engineering