Documentation Status

velodyne: velodyne_driver | velodyne_msgs | velodyne_pointcloud

Package Summary

Documented

Point cloud conversions for Velodyne 3D LIDARs.

velodyne: velodyne_driver | velodyne_msgs | velodyne_pointcloud

Package Summary

Documented

Point cloud conversions for Velodyne 3D LIDARs.

velodyne: velodyne_driver | velodyne_msgs | velodyne_pointcloud

Package Summary

Documented

Point cloud conversions for Velodyne 3D LIDARs.

velodyne: velodyne_driver | velodyne_msgs | velodyne_pointcloud

Package Summary

Released Continuous integration Documented

Point cloud conversions for Velodyne 3D LIDARs.

velodyne: velodyne_driver | velodyne_msgs | velodyne_pointcloud

Package Summary

Released Continuous integration Documented

Point cloud conversions for Velodyne 3D LIDARs.

Cannot load information on name: velodyne_pointcloud, distro: jade, which means that it is not yet in our index. Please see this page for information on how to submit your repository to our index.

Overview

This package provides point cloud conversions for Velodyne 3D LIDARs. It supports all the HDL-64E models, and the HDL-32E.

The API review describes the evolution of these interfaces.

New in Indigo: the default ~min_range value is now 0.9 meters.

New in Indigo: a new pair of parameters ~view_direction and ~view_width may be used to reduce the output point cloud to a subset of angular directions. By default, every angle is included in the point cloud. Setting ~view_width to pi/2 radians will limit the output to 90 degrees around the forward direction of the device (from -45 degrees to +45). Also setting ~view_direction to pi would return output only from the device's rear facing, instead. Similarly, setting view_direction to -pi/2 would limit output to 90 degrees around the right facing in the XY plane of the device frame of reference.

ROS Nodes and Nodelets

CloudNodelet

This nodelet reads raw data from the velodyne_packets ROS topic, converts to sensor_msgs/PointCloud2 format, and republishes as velodyne_points in the original frame of reference (typically /velodyne). In addition to the XYZ points, this cloud includes fields for "intensity" and "ring".

Subscribed Topics

velodyne_packets (velodyne_msgs/VelodyneScan)
  • one or more raw Velodyne data packets from the device.

Published Topics

velodyne_points (sensor_msgs/PointCloud2)
  • accumulated Velodyne points transformed in the original frame of reference.

Parameters

~max_range (double, default: 130.0)
  • maximum range to publish
~min_range (double, default: New in Indigo 0.9, formerly 2.0)
  • minimum range to publish
~calibration (string)
  • YAML file containing device-specific calibration information.
~view_direction (double, default: 0.0)
  • New in Indigo: center of view angle in radians in the device frame of reference.
~view_width (double, default: 2 * pi)
  • New in Indigo: width of view angle in radians in the XY plane of the device frame of reference.

Examples

Continuously convert raw Velodyne packets into sensor_msgs/PointCloud2 messages. Running as a standalone nodelet prevents zero-copy message sharing.

$ rosrun nodelet nodelet standalone velodyne_pointcloud/CloudNodelet

This launch file runs the cloud nodelet in the same process with the device driver. That enables zero-copy message sharing. The full path name of the calibration file must be provided. This example uses one of the package test files for calibration.

<launch>
  <!-- start nodelet manager and driver nodelets -->
  <include file="$(find velodyne_driver)/launch/nodelet_manager.launch" />

  <!-- start cloud nodelet -->
  <include file="$(find velodyne_pointcloud)/launch/cloud_nodelet.launch">
    <arg name="calibration"
         value="$(find velodyne_pointcloud)/params/64e_utexas.yaml"/>
  </include>

</launch>

cloud_node

This node runs the same code as CloudNodelet in a node by itself.

Examples

Continuously convert raw Velodyne packets into sensor_msgs/PointCloud2 messages.

$ rosrun velodyne_pointcloud cloud_node _calibration:=calibration.yaml

TransformNodelet

This nodelet reads raw data from the velodyne_packets ROS topic, converts to sensor_msgs/PointCloud2 format, and republishes as velodyne_points in a specified frame of reference (typically /odom). In addition to the XYZ points, this cloud includes fields for "intensity" and "ring".

Subscribed Topics

velodyne_packets (velodyne_msgs/VelodyneScan)
  • one or more Raw Velodyne data packets from the device

Published Topics

velodyne_points (sensor_msgs/PointCloud2)
  • accumulated Velodyne points transformed into the frame_id frame of reference.

Parameters

~frame_id (str, default: "odom")
  • Target frame ID. Resolved using tf_prefix, if defined.
~max_range (double, default: 130.0)
  • maximum range to publish
~min_range (double, default: New in Indigo 0.9, formerly 2.0)
  • minimum range to publish
~calibration (string)
  • YAML file containing device-specific calibration information.
~view_direction (double, default: 0.0)
  • New in Indigo: center of view angle in radians in the device frame of reference.
~view_width (double, default: 2 * pi)
  • New in Indigo: width of view angle in radians in the XY plane of the device frame of reference.

Examples

Transform raw Velodyne packets into sensor_msgs/PointCloud2 messages into the /odom frame. Running as a standalone nodelet prevents zero-copy message sharing.

$ rosrun nodelet nodelet standalone velodyne_pointcloud/TransformNodelet

This launch file runs the transform nodelet in the same process with the device driver. That enables zero-copy message sharing. The full path name of the calibration file must be provided. This example uses one of the package test files.

<launch>
  <!-- start nodelet manager and driver nodelets -->
  <include file="$(find velodyne_driver)/launch/nodelet_manager.launch" />

  <!-- start transform nodelet -->
  <include file="$(find velodyne_pointcloud)/launch/transform_nodelet.launch">
    <arg name="calibration"
         value="$(find velodyne_pointcloud)/params/64e_utexas.yaml"/>
  </include>
</launch>

transform_node

This node runs the same code as TransformNodelet in a node by itself.

Examples

Transform raw Velodyne packets into sensor_msgs/PointCloud2 messages into the /odom frame.

$ rosrun velodyne_pointcloud transform_node _calibration:=calibration.yaml

Transform raw Velodyne packets into sensor_msgs/PointCloud2 messages into the /map frame.

$ rosrun velodyne_pointcloud transform_node _frame_id:=/map

Launch File Examples

In two separate terminal windows, start a velodyne_nodelet_manager process running the driver nodelet and the cloud nodelet, which will have zero-copy access to the raw data messages the driver publishes.

 $ roslaunch velodyne_driver nodelet_manager.launch 

 $ roslaunch velodyne_pointcloud cloud_nodelet.launch calibration:=~/mydata.yaml

Start a driver nodelet with input from tcpdump.pcap, in the current directory. The pwd provides a full path name, as required for roslaunch. In another terminal, start the transform nodelet, to publish the data points transformed into the /odom frame of reference.

 $ roslaunch velodyne_driver nodelet_manager.launch pcap:=$(pwd)/tcpdump.pcap 

 $ roslaunch velodyne_pointcloud transform_nodelet.launch calibration:=~/mydata.yaml

Start a velodyne_nodelet_manager process for a Velodyne HDL-32E. This script runs both the driver and the point cloud conversion, providing the standard HDL-32E calibration.

 $ roslaunch velodyne_pointcloud 32e_points.launch

Start another velodyne_nodelet_manager for the Velodyne HDL-32E, but provide a PCAP dump file as input.

 $ roslaunch velodyne_pointcloud 32e_points.launch pcap:=$(pwd)/tcpdump.pcap

Utility Commands

gen_calibration.py

This script generates a YAML calibration file for use by this package from the db.xml file that was provided by Velodyne with the device.

Examples

Read db.xml from the current directory, writing the required calibration data to db.yaml.

 $ rosrun velodyne_pointcloud gen_calibration.py db.xml

Save generated 32E calibration data in my_calibration.yaml.

 $ rosrun velodyne_pointcloud gen_calibration.py 32db.xml my_calibration.yaml 

C++ API

C++ interfaces exported by this package:

Wiki: velodyne_pointcloud (last edited 2014-08-16 16:14:05 by JackOQuin)