Development and Testing of a Computer Vision based Indoor Positioning System for an upcoming Spacecraft Close-Proximity Operations test facility
- Paper number
GLEX-2025,13,1,5,x93212
- Author
Ms. Nitika Jaggi, Indian Institute of Technology Kanpur, India
- Coauthor
Mr. Shanmukha Sree Mokkapati, Indian Institute of Technology Kanpur, India
- Coauthor
Mr. Kiran Kumar Das, Lovely Professional University, India
- Coauthor
Mr. Sayandev Som, Indian Institute of Technology Kanpur, India
- Coauthor
Dr. Dipak Kumar Giri, Indian Institute of Technology Kanpur, India
- Year
2025
- Abstract
Spacecraft close-proximity operations are vital for space exploration, defense, and scientific missions. To ensure mission success, Guidance Navigation and Control (GNC) algorithms for close-proximity operations must be rigorously tested on the ground before deployment. Many space agencies employ ground-based spacecraft maneuvering control simulators that mimic the conditions their algorithms would face in space. The mock spacecraft testbeds move in a controlled manner in an arena where experiments are performed. The testbeds in such a setup need accurate localization information within this operational environment and therefore, an Indoor Positioning System is required. While testbeds in the past have used various types of motion capture systems, they are expensive and overkill for the task at hand. With the availability of open-source Computer Vision algorithms for specially designed fiducial markers like ArUco, a much simpler, relatively inexpensive, and computationally fast solution for navigation can be built. Through this paper, an ArUco marker and Computer Vision based navigation system for a 3-DOF spacecraft close-proximity operations testbed is being proposed. The mock spacecraft would translate in the XY plane within a square arena of 3m side. Each testbed would have a separate ID marker fixed to its top face. Two downward-facing cameras mounted on the ceiling above the arena would detect and track these tags. To characterize the performance of the navigation system, two wheeled robots would be used as surrogates for the spacecraft. An ArUco marker placed in the center of the arena would act as the origin of the inertial reference frame. The data from cameras would be combined using a complementary filter to estimate the robots' position relative to the origin. Orientation (yaw angle) estimation would be carried out by fusing data from the cameras and an onboard gyroscope, using a Kalman filter, implemented in MATLAB. To ensure accurate navigation performance, the arena would be placed in a controlled dark room with lights positioned at carefully chosen points. This would help to avoid shadow formation that could adversely affect marker detection. Additionally, since this test facility would be used to develop relative navigation algorithms in the future, to simulate more realistic scenarios, a spotlight to represent the sun and a projector to display Earth orbit images as the background, would be added. The proposed indoor positioning system would offer reliable position and velocity feedback of the mock satellites. This setup would enable future advancements in GNC algorithms for spacecraft proximity operations.
- Abstract document
- Manuscript document
GLEX-2025,13,1,5,x93212.pdf (🔒 authorized access only).
To get the manuscript, please contact IAF Secretariat.