Analysis and Design of a Sub-Optimal MEKF for Low Earth Orbit Attitude Estimation Using a Radically Inexpensive MEMS IMU

Session

Pre-Conference Workshop Session VI: Advanced Concepts III

Location

Utah State University, Logan, UT

Abstract

A previous study investigated the feasibility in using a radically inexpensive MEMS IMU with a GPS and a model of Earth’s magnetic field for attitude determination. A Multiplicative Extended Kalman Filter was designed to estimate the biases and errors associated with the IMU and reduce attitude uncertainty. States with large influence on overall uncertainty were identified through error budget and sensitivity analysis. It was determined that the complexity of the Kalman filter could be significantly reduced by removing 18 of the 42-element state vector. In this study, the sub-optimal filter is designed and its feasibility for attitude determination is demonstrated through Monte Carlo simulations. The primary mission is inertial attitude control in support of spacecraft mission operations in low Earth orbit.

Our challenge is to create a radically inexpensive spacecraft using commercially available and bespoke subsystems. This inhibits the use of expensive and larger attitude sensors such as star cameras and sun sensors. While the proposed system uses a Raspberry Pi as the main flight computer, reducing computational complexity enhances the ability to provide near-real-time attitude solutions. This sub-optimal MEKF therefore improves mission capabilities by reducing computational load.

This document is currently not available here.

Share

COinS
 
Aug 1st, 12:00 AM

Analysis and Design of a Sub-Optimal MEKF for Low Earth Orbit Attitude Estimation Using a Radically Inexpensive MEMS IMU

Utah State University, Logan, UT

A previous study investigated the feasibility in using a radically inexpensive MEMS IMU with a GPS and a model of Earth’s magnetic field for attitude determination. A Multiplicative Extended Kalman Filter was designed to estimate the biases and errors associated with the IMU and reduce attitude uncertainty. States with large influence on overall uncertainty were identified through error budget and sensitivity analysis. It was determined that the complexity of the Kalman filter could be significantly reduced by removing 18 of the 42-element state vector. In this study, the sub-optimal filter is designed and its feasibility for attitude determination is demonstrated through Monte Carlo simulations. The primary mission is inertial attitude control in support of spacecraft mission operations in low Earth orbit.

Our challenge is to create a radically inexpensive spacecraft using commercially available and bespoke subsystems. This inhibits the use of expensive and larger attitude sensors such as star cameras and sun sensors. While the proposed system uses a Raspberry Pi as the main flight computer, reducing computational complexity enhances the ability to provide near-real-time attitude solutions. This sub-optimal MEKF therefore improves mission capabilities by reducing computational load.