A Combined Filter for GPS-Based Attitude and Baseline Estimation

Publisher Website
Google Scholar
Abstract
This paper describes a methodology for combining attitude and baseline estimation into one comprehensive filter. The nominal filter states include attitude quaternion errors, angular velocity errors, and baseline errors. Previous approaches have relied on determining accurate baselines first and then using those baselines for attitude estimation. The method presented here combines these two steps into one algorithm. The advantage is that baseline calibration can be performed while simultaneously producing attitude information. In addition, the combined filter allows for baseline flexure, whereas previous approaches are based on the assumption of fixed baselines. This algorithm works best when attitude motion is present and when a good dynamic model is available. Results from simulated data as well as flight data are presented.

This publication has 1 reference indexed in Scilit: