Pointcloud_to_laserscan for bag files?

Question

Hello all,

I am currently working to analyze pointcloud data from a Velodyne LIDAR and run SLAM on it, so I've been trying to run pointcloud_to_laserscan on it in order to convert it into a form that many SLAM packages will accept. The data I'm working with is inside a bag file.

However, I'm having a number of issues getting this to work. I can run a pointcloud_to_laserscan node, but the thing doesn't seem to be subscribing to anything and thus I can't get my pointcloud data to it. I'm new to ROS and thus have a lot of trouble understanding the Kinect-based launch file approach since I'm trying to pull from a bag file rather than trying to read from a camera. Does anyone have any advice on how to handle this situation? Know any free 3D SLAM packages for Velodyne? Am I just stupid and there's an obvious, easy way to go about this that I'm missing?

Edit: I've been working on getting Pointcloud_to_laserscan working, somewhat have launch files down. The issue I'm getting now is that this module seems to be subscribing to the /tf and /tf_static topics instead of the pointcloud one. My current launch file is below.

<launch>

  <arg name="path" default="/media/sf_Research/Data-2016-03-16-LidarMBI-Bag/" />
  <arg name="file_1" default="12-L-shape-in-parkinglot.bag" />
  <arg name="file_2" default="16-back-to-building-along-walkway.bag" />

  <node pkg="pointcloud_to_laserscan" type="pointcloud_to_laserscan_node" name="pointcloud_to_laserscan">
<remap from="cloud_in" to="velodyne_points"/>
<rosparam>
            # target_frame: camera_link # Leave disabled to output scan in pointcloud frame
            transform_tolerance: 0.01
            min_height: 0.0
            max_height: 1.0

            angle_min: -1.5708 # -M_PI/2
            angle_max: 1.5708 # M_PI/2
            angle_increment: 0.087 # M_PI/360.0
            scan_time: 0.3333
            range_min: 0.45
            range_max: 4.0
            use_inf: true

            # Concurrency level, affects number of pointclouds queued for processing and number of threads used
            # 0 : Detect number of cores
            # 1 : Single threaded
            # 2->inf : Parallelism level
            concurrency_level: 1
        </rosparam>
  </node>

  <node pkg="rosbag" type="play" name="player" output="screen" args="--clock $(arg path)$(arg file_2).bag"/>

</launch>

Double Edit: The solution below works perfectly for what I need to do, thank you so much to kmhallen for sending it. Any future people who have the same problem, use the link below.

---

Originally posted by Autumneverlast on ROS Answers with karma: 3 on 2016-06-01

Post score: 0

Answer 1

http://wiki.ros.org/but_velodyne

https://github.com/robofit/but_velodyne

This might be useful. I don't know much about it.

---

Originally posted by kmhallen with karma: 1416 on 2016-06-02

This answer was ACCEPTED on the original site

Post score: 2

---

Original comments

Comment by Autumneverlast on 2016-06-03:
The package as uploaded had many, many issues, but once those were fixed (through downloading dependencies and updating some files), it worked. It worked like a charm. Thank you so much, you're an absolute life saver.

Out of curiosity, how did you find this package?

Comment by kmhallen on 2016-06-03:
A co-worker used it to extract a single scan from the large pointcloud.

Were you able to build a 3D map and localize with just a velodyne point cloud? If so, this is the only open source 3D SLAM implementation I am aware of. Can you post results?

Comment by Autumneverlast on 2016-06-06:
That's wonderful, if it's not too much trouble I would appreciate it if you could thank them on my behalf.

And I wasn't able to build a 3D map and localize with just the pointcloud data, I'm currently working to add odometry data via output from the robot in order to add localization for SLAM.