0
$\begingroup$

Rosanswers logo

Hi there,

I am running hector_slam in ROS Hydro with a Neato XV-11 laser and a Razor 9dof IMU.

The IMU has been included via hector_imu_attitude_to_tf.

In RViz everything looks good, I have a base_footprint frame that only moves on X and Y (and yaw) and attached to it a base_link frame that has additional roll and pitch movement.

Now my question is this:

Right now the IMU data is only transforming a frame but I guess that has no direct effect on the computation of hector_slam? Do I have to specify somewhere that hector_slam should consider the IMU data to increase its own accuracy? Or do I have to run an additional node for that, like hector_localization?


Originally posted by tinytron on ROS Answers with karma: 45 on 2014-12-01

Post score: 0

$\endgroup$

1 Answer 1

0
$\begingroup$

Rosanswers logo

hector_slam only uses (roll/pitch) imu data implicitly by transforming the scan through tf. It would be possible to for example use IMU data for improving the scan-matching start estimate, but this is currently not implemented.


Originally posted by Stefan Kohlbrecher with karma: 24361 on 2014-12-02

This answer was ACCEPTED on the original site

Post score: 1


Original comments

Comment by tinytron on 2014-12-02:
Ok thanks!

Comment by wyasuda on 2015-01-01:
Dear Stefan, I have a question about this issue. I read the pdf below. http://tedusar.eu/cms/sites/tedusar.eu.cms/files/Hector_SLAM_USAR_Kohlbrecher_RRSS_Graz_2012.pdf On page 32, you explained the Handheld Mapping System has IMU. Does IMU has no direct effect on the computation of hector_slam?

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.