ifm3d-ros2 overview

This documentation is formatted to be read on www.ros2.ifm3d.com.

:::{note} This release is intended to be used with the O3R camera platform ONLY. For other ifm cameras (e.g. O3D3xx and O3X1xx) please see the tagged releases 0.3.0 and 0.7.0 respectively. :::

ifm3d-ros2 is a wrapper around ifm3d enabling the usage of ifm O3R camera platform from within ROS 2 software systems.

As of version 1.2.0, ifm3d_ros2 also supports the Obstacle Detection Solution (ODS). Refer to the ODS documentation for more details on this application.

Software Compatibility Matrix

Releases versions

Note: The version numbers listed above for the ifm3d API or the O3R firmware versions are the ones explicitly tested. Any other version might work but is not officially supported.

Known Issues

For a complete list of changes, refer to the changelog.

Below is a list of all known issues with the latest released version:

• There is an issue in how the intrinsic parameters are calculated: see issue 18.
• The camera_info topic does not work for rectification: see issue 24.

Deprecated ifm3d-ros2 versions

The following versions are deprecated and no longer supported. | ifm3d_ros2 version | ifm3d version | ROS 2 distribution | | ——————– | ————- | —————— | | 1.0.1 DEPRECATED | 0.93.0 | Galactic | | 1.0.0 DEPRECATED | 0.92.0 | Galactic | | 0.3.0 DEPRECATED | 0.17.0 | Dashing, Eloquent | | 0.2.0 DEPRECATED | 0.12.0 | Dashing | | 0.1.1 DEPRECATED | 0.12.0 | Dashing | | 0.1.0 DEPRECATED | 0.12.0 | Dashing |

LICENSE

Please see the file called LICENSE.

\^\^\^\^\^\^\^\^\^\^\^\^\^\^\^\^\^\^\^\^\^\^\^\^\^ Changelog for package ifm3d-ros2 \^\^\^\^\^\^\^\^\^\^\^\^\^\^\^\^\^\^\^\^\^\^\^\^\^ 1.2 ===

1.2.0

• Create an ODS node to publish ODS data:
  • The launch file [ods.launch.py]{.title-ref} can be used,
  • Add two topics, ["~/ods_info"]{.title-ref} and ["~/ods_occupancy_map_ros"]{.title-ref},
  • An example launch configuration for ODS is provided [ods_default_parameters.yaml]{.title-ref} and can be used with the launch file,
  • It is expected that ODS is configured before the node is launched. Alternatively, one can use the new [config_file]{.title-ref} parameter.
• Remove the [diag_mode]{.title-ref} parameter: diagnostic is always polled periodically and published to ["/diagnostic"]{.title-ref}
• Add the [GetDiag]{.title-ref} service for polling filtered diagnostic data.
• Camera info topic:
  • Add a ["~/camera_info"]{.title-ref} topic for RGB cameras.
  • The ["~/camera_info"]{.title-ref} topic for the TOF cameras is published when the [TOF_INFO]{.title-ref} buffer is requested, instead of the [INTRINSIC]{.title-ref} buffer.
• Add a [config_file]{.title-ref} parameter. It should be formatted in JSON and will be used to configure the device when the [CONFIGURE]{.title-ref} state is triggered.

* Transforms:

• The [cloud_link]{.title-ref} was renamed to [ifm_base_link]{.title-ref}, and is used as the reference ifm calibrated coordinate system for all ifm data (RGB, 3D and ODS).
• The transforms between the [cloud_link]{.title-ref} (now [ifm_base_link]{.title-ref}), and the [mounting_link]{.title-ref} and [optical_link]{.title-ref} are fixed.
• The [mounting_link]{.title-ref} to [optical_link]{.title-ref} transform is read when the first [TOF_INFO]{.title-ref} or [RGB_INFO]{.title-ref} buffer is received, and remains constant.
• The [ifm_base_link]{.title-ref} to [mounting_link]{.title-ref} transform is published once when the first [TOF_INFO]{.title-ref} or [RGB_INFO]{.title-ref} buffer is received, and is only re-published subsequently if changed.
• The camera node parameters related to tf publication got reworked, the new parameters are:
  • `tf.base_frame_name`: Name for ifm reference frame

|   * `tf.mounting_frame_name`: Name for the mounting point frame |   * `tf.optical_frame_name`: Name for the optical frame |   * `tf.publish_base_to_mounting`: Whether the transform from the ifm base link to the camera mounting point should be published |   * `tf.publish_mounting_to_optical`: Whether the transform from the cameras mounting point to the optical center should be published

1.1

1.1.0 (unreleased preparations for FW 1.4 release) ------------------* Update for compatibility with O3R FW 1.4.x and ifm3d >= 1.4.3 * Tested with FW 1.4.22 and ifm3d API 1.4.3 * Tested with FW 1.4.22 and ifm3d API 1.5.3

1.1.0

• Update for compatibility with O3R FW >= 1.0.14 and ifm3d >= 1.2.6
  • Using ifm3d::O3R instead of ifm3d::Device to access camera
  • Added [buffer_id_list]{.title-ref} param to define which data to get from the device and consequently which topics to publish. Maps to ifm3d [buffer_id]{.title-ref}.
• Remove the need to use cyclonedds: TODO: verification
• Remove dependency to Boost library
• Include Dockerfile and helper scripts:
  • Dockerfile for building the ROS2 node inside a Docker container
  • Helper build and launch scripts for ROS Docker images
  • Adds docker-compose example files for launching the ROS node on a O3R edge device: VPU
• Moved Publishers into node namespace: This matches the ifm3d-ros2 node version's naming (<= 1.0.x).
• Switch to a new camera.launch.py launch file (instead of camera_managed and camera_standalone): launches and activates a single camera node
  • focusses on configurability
  • camera_default_parameters.yaml contains only default params
  • Add option for visualization to launch file: If visualization is true, RViz2 with the config file from etc is opened
  • Formatting of node names improved
• old launch files are marked as deprecated and will be removed with next release: please migrate your launch files to the new structure
• Switching json handling to ifm3d::json
• Reduce logging level for continuous outputs.
• Added topics:
  • diagnostics (published to global /diagnostic topic),
  • camera_info,
  • TOF_INFO,
  • RGB_INFO,
  • extrinsics
• Added parameters:
  • log_level,
  • tf related parameters (see parameters doc for more details),
  • buffer_id_list,
  • diag_mode

1.0

1.0.4 (unreleased)

• Updating to underlying ifm3d API version 1.1.1
• Name change for TOF_INFO and RGB_INFO buffer.
• Added Error, AsynError and AsyncNotification callbacks.

1.0.3 (unreleased)

• Updating to underlying ifm3d API version 1.0.1
• Switching to using buffer_ids instead of schemas to determine data types.
• Introducing buffer_id_utils.hpp for buffer_id handling.

1.0.2

• Fixed tf chain in launchfiles

1.0.1

• Removed dependencies to OpenCV and PCL and the Image module of the ifm3d library. Now using the StlImage module of ifm3d.
• The RGB image is published as a compressed JPEG image. To view it with RViz, you can use the image_transport republish node (see documentation).
• Update the launchfiles for compatibility with pyyaml 6.0 (backward compatibility maintained down to pyyaml 5.1).

1.0.0

• Update the ROS node for the O3R camera platform
• add a 2D RGB data publisher
• add support for specifying the PCIC data communication TCP port
• add example yaml files and launch file for multi PCIC TCP port settings
• add service functions for setting each camera head to RUN / IDLE state
• update the dump and config service to O3R JSON structure
• removed unit vector publishing: please use the intrinsic camera calibration parameters instead
• removed the temperature publisher until diagnose chunk is available
• removed axis permutation for ROS node specific axis order
• removed xyz image publisher

known limitations

• DDS settings need to be set to cyclonedds - ROS 2 Galactic is the preferred ROS 2 distro atm.

0.3.0 (2020-01-31) --------

• Publish camera temperature (Thanks \@dustingooding of Houston Mechatronics)
• Tested on Eloquent

0.2.0 (2019-06-26) --------

• Provided an implementation of the [Dump]{.title-ref} service
• Provided an implementation of the [dump]{.title-ref} command-line tool
• Provided an implementation of the [Config]{.title-ref} service
• Provided an implementation of the [config]{.title-ref} command-line tool

0.1.1 (2019-06-25) --------

• Unit vectors are published on a "latched topic". Please see [qos.hpp]{.title-ref} for the specifics of the QoS profile [ifm3d_ros2::LatchedQoS]{.title-ref}.
• On-camera extrinsics in support of off-line point cloud computation are now exposed.
• Bootstrapped some unit tests
• Cleaned up some documentation.

0.1.0 (2019-06-06) --------

• Initial (alpha) release