A Low-Cost Air Bearing based ADCS Calibration System

A flexible and reconfigurable Attitude Determination and Control System (ADCS) testbed is essential for hardware-in-the-loop (HIL) testing and performance verification of small satellites. This paper presents a low-cost and reconfigurable vision-based motion capture system that enables accurate tracking of rotational motion on an air-bearing testbed. The system employs multiple LEDs on air-bearing testbed as feature points, and two or more webcams as the vision-based motion capture platform. The LEDs are detected and tracked using the You-Only-Look-Once (YOLO) algorithm. An extended Kalman filter (EKF) fuses the detected measurements to estimate both the rotation rate and LED positions in real time. Monte Carlo simulations demonstrate that an accuracy of 0.1 deg/s is achievable under varying error magnitudes. Preliminary experiments confirm the feasibility and stability of the proposed method, even with low-cost cameras and without precise calibration, validating its suitability for reconfigurable ADCS testbeds.