Master's Thesis Proposal

Kevin Zhang

(Prof. John Christian)

Optimizing Camera Calibration Operations using Star Fields to Enable Precision Optical Navigation

Friday, Nov. 21

10:00 a.m.

CODA 0915

Abstract:

Camera calibration is essential for the use of precise vision navigation algorithms in space. While preflight calibration of optical instruments typically occurs, experience has shown that calibration parameters change on orbit. Nearly every contemporary mission relying on vision navigation algorithms performs in-orbit camera calibration by opportunistically capturing star field images. This practice neglects the quality of images collected, which is experimentally confirmed to yield less informative measurements. An optimal design of experiments approach for capturing image sequences of stars is formulated while considering operational mission constraints. Furthermore, although camera calibration is well studied for narrow field-of-view (FOV) cameras, the routine is insufficient when applied for wide FOV cameras. Wide FOV cameras are effective for terrain relative navigation and hazard detection but suffer from severe and complex distortions. Current methods of achieving a precise calibration tend to be manual and heuristic. An automated framework is proposed for data-driven lens model selection and star correspondence extraction. These contributions reduce mission operation costs and, ultimately, provide higher quality camera distortion models necessary for precision navigation.

Committee:
Prof. John Christian (advisor), School of Aerospace Engineering
Prof.  E. Glenn Lightsey, School of Aerospace Engineering
Prof. Koki Ho, School of Aerospace Engineering
Courtney Mario, Draper
,