Recursive Inertia Estimation With Semidefinite Programming - Robotics Institute Carnegie Mellon University

Recursive Inertia Estimation With Semidefinite Programming

Zac Manchester and Mason Peck
Conference Paper, Proceedings of AIAA Guidance, Navigation, and Control Conference (GNC '17), January, 2017

Abstract

A recursive algorithm suitable for on-orbit estimation of a spacecraft’s inertia tensor is presented. The estimation problem is formulated as a semidefinite program with explicit enforcement of positive definiteness and other constraints on the elements of the inertia tensor, which enhances convergence and guarantees that a physically valid result is returned. A discrete mechanics formulation of the spacecraft dynamics allows data routinely generated by spacecraft attitude determination systems to be directly used as input to the estimator. Numerical examples demonstrate the performance of the
algorithm on simulated telemetry data.

BibTeX

@conference{Manchester-2017-122119,
author = {Zac Manchester and Mason Peck},
title = {Recursive Inertia Estimation With Semidefinite Programming},
booktitle = {Proceedings of AIAA Guidance, Navigation, and Control Conference (GNC '17)},
year = {2017},
month = {January},
}