In a Twist message, how should angular velocity be interpreted?

asked 2021-04-22 09:23:38 -0600

351 ●9 ●14 ●16

Is the angular velocity instead a vector whose magnitude is the speed of rotation (in rad/s), and whose direction is the axis of rotation? Or is it something different, like Euler angular velocity (unless those are equivalent)?

I tried to look at https://en.wikipedia.org/wiki/Screw_t... which was linked elsewhere, but it sounds like it describes something different, where rotation can only happen around the axis along which it moves.

edit retag flag offensive close merge delete

add a comment

1

answered 2021-05-05 12:01:14 -0600

nnmm

351 ●9 ●14 ●16

I'm fairly sure now that both interpretations are equivalent. The beginning of this video says this too: "[The components] are the instantaneous body rates about the [X, Y, Z] axes".

edit flag offensive delete link

add a comment

1

answered 2021-04-22 12:19:55 -0600

shonigmann

1567 ●15 ●115 ●49

I believe #9941 will answer your question, but please follow up if its not clear.

In short, the linear and angular components of twist messages represent the instantaneous linear and angular velocities. In most applications (that I have been exposed to), it is common to use the robot's body frame (e.g. base_link) as the default reference frame, because this is typically the frame you have control over. If you need that to be in a specific reference frame, you can use TwistStamped messages, which include the frame in the header.

edit flag offensive delete link

add a comment

In a Twist message, how should angular velocity be interpreted?

2 Answers

Question Tools

Stats

Related questions

In a Twist message, how should angular velocity be interpreted? edit

2 Answers

Question Tools

Stats

Related questions

In a Twist message, how should angular velocity be interpreted?