0
$\begingroup$

Rosanswers logo

I am trying my hand at robot navigation. I am not able to comprehend, the following differences,

  1. Difference in the odometry data from sources like wheel or visual and the output generated by state estimators like robot_pose_ekf packages. Because they both seem to have the same message format.

  2. I am able to understand the transform between baselink and different sensor coordinate frames. But what is the difference between world, map, nav, odom and base_link frame. In some cases I find, base_footprint and base_stabilized coordinate frames too.

  3. In the tf map, should there be one fixed frame and can odom_frame be made as fixed frame in the presence of /map or /world frame?


Originally posted by boon on ROS Answers with karma: 111 on 2015-09-26

Post score: 0

$\endgroup$

1 Answer 1

0
$\begingroup$

Rosanswers logo

  1. Data from sources like wheel encoders or visual odometry can be noisy and might not be good enough. Therefore, robot_pose_ekf combines measurements coming from different sources (like IMU, Wheel encoders, Visual Odometry) using EKF to provide a better estimate of the 3D pose of the robot.

  2. REP-105 explains different coordinate frames. base_footprint is just a projection of base_link on the ground.

  3. Yes.. odom_frame can be made a fixed frame in presence of other frames. I am not 100% sure about this. Can someone confirm this?

Hope it helps.


Originally posted by Naman with karma: 1464 on 2015-09-26

This answer was ACCEPTED on the original site

Post score: 0

$\endgroup$

Your Answer

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