kinect2-ros

Synopsis

Tools for using the Kinect One/v2 in ROS.

Author(s): Thiemo Wiedemeyer, Ralf Kaestner

Maintainer: Ralf Kaestner [email protected]

License: Apache License

Operating system(s): Debian-based Linux

Package PPA: ppa:ethz-asl/ros

Description

This project contains a collection of tools and libraries for using the Kinect One/v2 in ROS. In particular, it provides:

• A calibration tool for calibrating the IR sensor of the Kinect One/v2 to the RGB sensor
• A library for depth registration with OpenCL support
• The bridge between libfreenect2 and ROS
• A viewer for the images/point clouds

This project has been forked from the original Kinect One/v2 ROS integration published under http://github.com/code-iai/iai_kinect2 in order to maintain Debian packaging and some feature additions.

Publications

If you use this work in an academic context, please cite the following publication:

P. Fankhauser, M. Bloesch, D. Rodriguez, R. Kaestner, M. Hutter, and R. Siegwart, "Kinect v2 for Mobile Robot Navigation: Evaluation and Modeling", in IEEE International Conference on Advanced Robotics (ICAR), 2015. (submitted) (PDF Draft)

@inproceedings{Fankhauser2015KinectV2ForMobileRobotNavigation,
  author = {Fankhauser, Péter and Bloesch, Michael and Rodriguez, Diego and and Kaestner, Ralf and Hutter, Marco and Siegwart, Roland},
  title = {Kinect v2 for Mobile Robot Navigation: Evaluation and Modeling},
  booktitle = {IEEE International Conference on Advanced Robotics (ICAR) (submitted)},
  year = {2015}
}

Installation

Here, we assume you intend to build/install the project for the ROS distribution named ROS_DISTRO.

Installing from packages (recommended for Ubuntu LTS users)

The maintainers of this project provide binary packages for the latest Ubuntu LTS releases and commonly used system architectures. To install these packages, you may follow these instructions:

• Add the backports PPA and the drivers PPA to your APT sources by issuing

  sudo add-apt-repository ppa:ethz-asl/backports
  sudo add-apt-repository ppa:ethz-asl/drivers
  

  on the command line

• Add the project PPA to your APT sources by issuing

  sudo add-apt-repository ppa:ethz-asl/ros
  

  on the command line

• To re-synchronize your package index files, run

  sudo apt-get update
  

• Install all project packages and their dependencies through

  sudo apt-get install ros-ROS_DISTRO-kinect2
  

  or selected packages using your favorite package management tool

Building from source

This project may be built using the CMake build system with the ROS macro extension called catkin.

Installing build dependencies

Most build dependencies of this project are available from the standard package repositories of recent Ubuntu and ROS releases. To install them, simply use the command

sudo apt-get install libopencv-de libpcl-1.7-all-dev ros-ROS_DISTRO-roscpp, ros-ROS_DISTRO-rostime, ros-ROS_DISTRO-std-msgs, ros-ROS_DISTRO-sensor-msgs, ros-ROS_DISTRO-message-filters, ros-ROS_DISTRO-cv-bridge, ros-ROS_DISTRO-image-transport, ros-ROS_DISTRO-compressed-image-transport, ros-ROS_DISTRO-compressed-depth-image-transport

The Kinect v2 driver library (and its dependencies) may conveniently be installed from the package PPAs of the project maintainers. To install them, you may follow these instructions:

• Add the backports PPA and the drivers PPA to your APT sources by issuing

  sudo add-apt-repository ppa:ethz-asl/backports
  sudo add-apt-repository ppa:ethz-asl/drivers
  

  on the command line

• To re-synchronize your package index files, run

  sudo apt-get update
  

• Install the required dependencies through

  sudo apt-get install libfreenect2-dev
  

  or using your favorite package management tool

Building with catkin (recommended for ROS developers)

Assuming that you have cloned the project sources into PROJECT_DIR, you may attempt to build this project using catkin as follows:

• Create the directory structure for your catkin workspace by issuing

  mkdir -p CATKIN_WS_DIR/src
  

• Assuming that your ROS environment has been set up properly, initialize the catkin workspace using the command

  catkin_init_workspace CATKIN_WS_DIR/src
  

• Individually link the directories containing the project's package source trees into your catkin workspace source directory through

  ln -sf PROJECT_DIR/kinect2* CATKIN_WS_DIR/src
  

• Switch into the catkin workspace directory by

  cd CATKIN_WS_DIR
  

• In the catkin workspace directory, run

  catkin_make
  

  to start the build

Usage

roslaunch kinect2_bridge kinect2_bridge.launch

There are several options to configure the kinect2_bridge. These can be set as argument, for example

roslaunch kinect2_bridge kinect2_bridge.launch publish_tf:=true fps_limit:=10

Commonly used options are:

• publish_tf (<bool>, default: false)

  Set to true to publish the tf frames.

• fps_limit (<float>, default: -1)

  Limiting the frame rate (frames per second). Set to -1 for unlimited.

• scale (<float>, default: -1)

  If greater than zero, apply scale to yield scaled color and depth image topics. Otherwise, scaled color and depth image topics have the native resolution of each image stream.

• calib_path (<string>, default: "$(find kinect2_bridge)/data/")

  Path to calibration files.

• depth_method (default|cpu|opengl|opencl, default: default)

  Selects the depth processing method.

• reg_enabled (<bool>, default: true)

  Set to true to activate depth/color registration.

• reg_method (default|cpu|opencl, default: default)

  Selects the depth/color registration method.

• reg_cl_source (<string>, default: "$(find kinect2_depth_registration)/opencl/depth_registration.cl")

  Path to the OpenCL program for depth/color registration.

For common pitfalls, check out this list from libfreenect2.

API documentation

This project does not yet provide any API documentation.

Feature requests and bug reports

If you would like to propose a feature for this project, please consider contributing or send a feature request to the project authors. Bugs may be reported through the project's issue page.

Further reading

For additional information of the Robot Operating System (ROS), please refer to the official ROS documentation.