Skip to content
Laughlin Barker edited this page Nov 8, 2017 · 3 revisions

This repository contains a Robot Operating System (ROS) project with a small library for testing the consistency of the Extended Kalman Filter (EKF).

This package provides two online statistical tests, which examine the statistics of the EKF innovation vector. Using the definition of filter consistency given in [1], namely a filter is consistent if:

  1. The state errors should be acceptable as zero mean and have magnitude commensurate with the state covariance as yielded by the filter.
  2. The innovations should also have the same property.
  3. The innovations should be accepted as white.

Implemented in this package is the time-average normalized innovation squared statistic, which computes

time-average normalized innovation squared statistic

where


[1] Bar-Shalom, Y., Li, X.R. and Kirubarajan, T., 2004. Estimation with applications to tracking and navigation: theory algorithms and software. John Wiley & Sons.

Clone this wiki locally