ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange
Ask Your Question
0

Several things on the laser_scan_matcher package tutorial

asked 2011-12-05 02:24:55 -0500

alfa_80 gravatar image

updated 2011-12-07 18:26:16 -0500

I was referring to a tutorial pertaining to laser_scan_matcher package. Thanks a lot for such a decent tutorial. Nevertheless, I have several questions regarding this package and the related written tutorial. They are as follow:

  1. It publishes pose2D topic which comprises of position(x, y) and orientation(theta). Are those position and orientation values cumulative or relative?
  2. Can the theta value be used without /imu and /odom? I actually got the theta values without /imu and /odom, are these values valid? I'm asking this because there is something written in the tutorial "Imu messages, used for theta prediction. Only used if use_imu is set to true.", so, I'm afraid that the valid theta values can only be obtained/valid when Imu messages is used.
  3. In the tutorial, "Required tf Transforms:base_link → laser". How do I implement this? Is it right/sufficient as what I did in the launch file here.

Hope somebody or the author can give me some insights on this..

edit retag flag offensive close merge delete

1 Answer

Sort by » oldest newest most voted
1

answered 2011-12-05 17:10:03 -0500

You are working with three frames: the fixed frame (let's say, odom), the base frame(base_link) and the laser frame (laser)

  1. The pose2D message is the same as the odom->base_link tf being published. It is provided only for convenience (in case you want to subscribe to a topic and not a tf). The value of the message is the pose of the base frame in relation to the fixed frame.

  2. Make sure you note the distinction between the theta input and theta output. The theta input is optional. It can be from an IMU or odometry source. If you have these sensors, it improves the speed and accuracy of the scan matching, because it provides an initial guess. The theta output is the best guess for the angle, given the laser scans. So yes, the theta output is valid even if you don't have any theta input.

  3. That tf best implemented with a static tf publisher. The launch file you linked works.

I hope this helps.

edit flag offensive delete link more

Comments

By the way, I only have 2 frames which are the base frame(base_link) and the laser frame (laser). I don't have the fixed frame (let's say, odom). Even without /odom, I still can subscribe /pose topic. Is that still OK, what I did?
alfa_80 gravatar image alfa_80  ( 2011-12-05 18:16:26 -0500 )edit
@Ivan Dryanovski: Could you please answer my question no.1 as well..Thanks..
alfa_80 gravatar image alfa_80  ( 2011-12-05 18:24:47 -0500 )edit

Question Tools

1 follower

Stats

Asked: 2011-12-05 02:24:55 -0500

Seen: 698 times

Last updated: Dec 07 '11