This paper presents a method of obtaining the maximum a posteriori estimate of an inspector satellite's trajectory about an unknown tumbling target while on-orbit. An inspector equipped with radar or a 3D visual sensor (such as LiDAR or stereo cameras), an inertial measurement unit, and a star tracker is used to obtain measurements of range and bearing to the target's centroid, angular velocity, acceleration, and orientation in the inertial frame. A smoothing-based trajectory estimation scheme is presented that makes use of all the input sensor data to estimate the inspector's trajectory. Open-source incremental smoothing and mapping (iSAM2) software is used to implement the smoothing-based trajectory estimation algorithm; this facilitates computationally efficient evaluation of the entire trajectory, which can be performed incrementally, and in real time on a computer capable of processing 3D visual sensor data in real time. The presented algorithm was tested on data obtained in 6 degree-of-freedom microgravity using the SPHERES-VERTIGO robotic test platform on the International Space Station (ISS). In these tests, a SPHERES inspector satellite with attached stereo cameras circumnavigated a passive SPHERES target satellite, making visual observations of it. The results of these tests demonstrate accurate estimation of the inspector satellite's trajectory.
Smoothing-based estimation of an inspector satellite trajectory relative to a passive object
2017-03-01
919302 byte
Conference paper
Electronic Resource
English
Faster Kalman trajectory smoothing
AIAA | 1977
|Trajectory Estimation for Satellite Clusters
Online Contents | 2006
|Trajectory Estimation for Satellite Clusters
AIAA | 2006
|Trajectory design for satellite relative angles-only navigation
American Institute of Physics | 2012
|