PhD Proposal by Christopher McBryde

Event Details
  • Date/Time:
    • Wednesday February 15, 2017
      12:00 pm - 2:00 pm
  • Location: Montgomery Knight 317
  • Phone:
  • URL:
  • Email:
  • Fee(s):
    N/A
  • Extras:
Contact
No contact information submitted.
Summaries

Summary Sentence: SPACECRAFT VISUAL NAVIGATION USING APPEARANCE MATCHING AND VISIBLE-INFRARED SENSOR FUSION

Full Summary: No summary paragraph submitted.

Doctoral Thesis Proposal

by

Christopher R. McBryde

Advisor: Prof. E. Glenn Lightsey

entitled

SPACECRAFT VISUAL NAVIGATION USING APPEARANCE MATCHING AND VISIBLE-INFRARED SENSOR FUSION

February 15  12 pm, Montgomery Knight 317

The last ten years have seen a fundamental change in the way spacecraft are designed and used. In the past, construction and launch of even a small satellite was only financially and logistically feasible for governments and large corporations. These devices were large, power-hungry, and needed to accomplish several tasks simultaneously in order to justify their cost. Since the rapid miniaturization of electronic components, the same capabilities can be implemented in a smaller, cheaper package that requires less power. Commercial-off-the-shelf hardware, like GPS receivers, magnetometers, and particularly cameras in both the visible and infrared wavelengths, is available to most consumers. Sensor fusion is therefore a more viable strategy for spacecraft sensing than it was 10 or 15 years ago. For object recognition and pose estimation, the appearance matching technique, introduced by Murase and Nayar in 1995, is a powerful algorithm that works particularly well with the on-orbit space environment.

Three contributions will be proposed: the spacecraft imaging simulation environment (SISE), the application of appearance matching to spacecraft object identification and pose estimation, the improvement of visual navigation through sensor fusion. The SISE will be a useful tool for software-in-the-loop verification and turning of the relative navigation algorithm. By employing appearance matching, training data can be generated for objects that cannot be imaged on the ground and return accurate relative pose information. Application of sensor fusion extends the usability of visual navigation into the eclipse portion of the orbit as well as augmenting the accuracy of object identification and pose estimation. The proposed research will result in an improvement in the accuracy and capability of spacecraft visual navigation sensing techniques.

Committee Members:

  • Dr. Glenn Lightsey, School of Aerospace Engineering (advisor)
  • Dr. Marcus Holzinger, School of Aerospace Engineering
  • Dr. Eric Johnson, School of Aerospace Engineering
  • Dr. Chad Frost, Ames Research Center
  • Dr. Andrew Johnson, Jet Propulsion Laboratory

Additional Information

In Campus Calendar
Yes
Groups

School of Aerospace Engineering, Graduate Studies

Invited Audience
Public
Categories
Other/Miscellaneous
Keywords
Phd proposal
Status
  • Created By: Margaret Ojala
  • Workflow Status: Published
  • Created On: Feb 1, 2017 - 3:38pm
  • Last Updated: Feb 8, 2017 - 11:40am