Sébastien Henry
(Advisor: Prof. Christian]

will propose a doctoral thesis entitled,

Absolute and Autonomous Spacecraft Navigation Using Line-of-Sight Measurements

On

Thursday, January 18 at 9:00 a.m.
Montgomery Knight Building 317
 

Abstract
Autonomous navigation in deep space will represent a key aspect of space missions in the coming decades. While many options exist, this proposal explores the use of spacecraft onboard cameras for optical navigation (OPNAV). Images of celestial objects combined with specific algorithms can allow a spacecraft to determine its attitude, position, and velocity. We pay particular attention to stars, unresolved objects in the solar system (e.g. planets, moons, asteroids), and specific feature keypoints on a planetary surface, where a single pixel coordinate can be extracted from the image as the measurement. This type of measurement represents a line-of-sight that contains information on the direction, but not the scale, from the camera to the objects. The proposal is structured around three contributions, incorporating concepts like astrodynamics, computer vision, special relativity, and much more.

First, we delve into triangulation, which is essential to find the position of the spacecraft when we observe at least two objects with known positions. We provide a new efficient triangulation algorithm and careful mathematical analysis of celestial effects to optimally localize the spacecraft. This contribution extends beyond the space domain as the mathematics apply to robotics and 3D reconstruction.

Second, we use line-of-sight measurements between two spacecraft augmented with inter-satellite range information in formation flying. Our algorithm allows absolute orbit determination for all spacecraft in a formation, without any prior information on the spacecraft states.

Third, we treat the management of line-of-sight navigation in a more systems engineering approach. Space images can be noisy or contain many different objects. We propose an image processing pipeline to match the correct object to the correct measurement in the image (e.g. identify a distant planet in a photo with stars in the background). These measurements can then feed a filter to complete the autonomous navigation pipeline.

Committee

• Prof. John A. Christian – School of Aerospace Engineering (advisor)
• Prof. Brian C. Gunter – School of Aerospace Engineering
• Prof. Edgar Glenn Lightsey – School of Aerospace Engineering