Attention: Answers.ros.org is deprecated as of August the 11th, 2023. Please visit robotics.stackexchange.com to ask a new question. This site will remain online in read-only mode during the transition and into the foreseeable future. Selected questions and answers have been migrated, and redirects have been put in place to direct users to the corresponding questions on Robotics Stack Exchange. Additional details are available here.

Why process noise covariance is multiplied by delta time when calculating estimate error covariance

asked 2020-09-11 00:45:46 -0600

forreg
1 ●1 ●1 ●1

I have noticed in both ekf and ukf, the cpp codes write

estimateErrorCovariance_.noalias() += delta * (*processNoiseCovariance);

What's the delta doing here?

edit retag flag offensive close merge delete

add a comment

1 Answer

Sort by » oldest newest most voted

answered 2020-09-11 12:57:00 -0600

JackB

977 ●10 ●76 ●33

Estimation filters are used to propogate the state of something through time. For probability distributions this means propogating the mean the covariance. In in order to do this we should multiply both the state and covariance by the time step by which we are predicting forward which is "delta".

edit flag offensive delete link

Comments

@forreg did this help answer your question or did you find the correct answer somewhere else and can share it here?

JackB ( 2020-09-21 21:24:23 -0600 )edit

add a comment

Question Tools

Stats

Asked: 2020-09-11 00:34:42 -0600

Seen: 175 times

Last updated: Sep 11 '20

Why process noise covariance is multiplied by delta time when calculating estimate error covariance edit

1 Answer

Comments

Question Tools

Stats

Related questions

Why process noise covariance is multiplied by delta time when calculating estimate error covariance