Documentation / Gemini 330 series /
ROS 2 Wrapper User Manual

ROS 2 Wrapper User Manual

stable version

Orbbec SDK ROS 2 is a wrapper for the Orbbec 3D camera that provides seamless integration with the ROS 2 environment. It
supports ROS 2 Foxy, Galactic, and Humble distributions.


Table of Contents


Installation Instructions

Install ROS 2

If your ROS 2 command does not auto-complete, put the following two lines into your .bashrc
or .zshrc

eval "$(register-python-argcomplete3 ros2)"
eval "$(register-python-argcomplete3 colcon)"

Create colcon workspace

mkdir -p ~/ros2_ws/src

Get source code

cd ~/ros2_ws/src
git clone

Install deb dependencies

# assume you have sourced ROS environment, same blow
sudo apt install libgflags-dev nlohmann-json3-dev libgoogle-glog-dev \
ros-$ROS_DISTRO-image-transport ros-$ROS_DISTRO-image-publisher ros-$ROS_DISTRO-camera-info-manager \
ros-$ROS_DISTRO-diagnostic-updater ros-$ROS_DISTRO-diagnostic-msgs

Install udev rules.

cd  ~/ros2_ws/src/OrbbecSDK_ROS2/orbbec_camera/scripts
sudo bash
sudo udevadm control --reload-rules && sudo udevadm trigger


Getting start

cd ~/ros2_ws/
# build release, Default is Debug
colcon build --event-handlers  console_direct+  --cmake-args  -DCMAKE_BUILD_TYPE=Release

Launch camera node

  • On terminal 1
. ./install/setup.bash
ros2 launch orbbec_camera # or other launch file, see below table
  • On terminal 2
. ./install/setup.bash

Select the topic you want to display

  • List topics / services/ parameters ( on terminal 3)
ros2 topic list
ros2 service list
ros2 param list
  • Get device info
ros2 service call /camera/get_device_info orbbec_camera_msgs/srv/GetDeviceInfo '{}'
  • Get SDK version
ros2 service call /camera/get_sdk_version orbbec_camera_msgs/srv/GetString '{}'

  • Get exposure
ros2 service call /camera/get_color_exposure orbbec_camera_msgs/srv/GetInt32 '{}'

If your check ir or depth, please change /camera/get_color_exposure
to /camera/get_ir_exposure or /camera/get_depth_exposure, Same below.

  • Get gain
ros2 service call /camera/get_color_gain orbbec_camera_msgs/srv/GetInt32 '{}'
  • Get white balance
ros2 service call /camera/get_white_balance orbbec_camera_msgs/srv/GetInt32 '{}'
  • Set auto exposure
ros2 service call /camera/set_color_auto_exposure std_srvs/srv/SetBool '{data: false}'
  • Set white balance
ros2 service call /camera/set_white_balance orbbec_camera_msgs/srv/SetInt32 '{data: 4600}'
  • Set laser enable
ros2 service call  /camera/set_laser_enable std_srvs/srv/SetBool "{data: true}"
  • toggle sensor
ros2 service call /camera/toggle_ir std_srvs/srv/SetBool "{data : true}"

  • save point cloud
ros2 service call /camera/save_point_cloud std_srvs/srv/Empty "{}"


Launch parameters

The following are the launch parameters available:

  • connection_delay: The delay time in milliseconds for reopening the device. Some devices, such as Astra mini, require
    a longer time to initialize and reopening the device immediately can cause firmware crashes when hot plugging.
  • enable_point_cloud: Enables the point cloud.
  • enable_colored_point_cloud: Enables the RGB point cloud.
  • point_cloud_qos, [color|depth|ir]_qos, [color|depth|ir]_camera_info_qos: ROS 2 Message Quality of Service (QoS)
    settings. The possible values
    case-insensitive. These correspond
    to rmw_qos_profile_system_default, rmw_qos_profile_default, rmw_qos_profile_parameter_events, rmw_qos_profile_services_default, rmw_qos_profile_parameters,
    and SENSOR_DATA, respectively.
  • enable_d2c_viewer: Publishes the D2C overlay image (for testing only).
  • device_num: The number of devices. This must be filled in if multiple cameras are required.
  • color_width, color_height, color_fps: The resolution and frame rate of the color stream.
  • ir_width, ir_height, ir_fps: The resolution and frame rate of the IR stream.
  • depth_width, depth_height, depth_fps: The resolution and frame rate of the depth stream.
  • enable_color: Enables the RGB camera.
  • enable_depth: Enables the depth camera.
  • enable_ir: Enables the IR camera.
  • depth_registration: Enables hardware alignment of the depth frame to the color frame. This field is required when
    the enable_colored_point_cloud is set to true.
  • usb_port: The USB port of the camera. This is required when multiple cameras are used.
  • enable_accel: Enables the accelerometer.
  • accel_rate: The frequency of the accelerometer, the optional values
    are 1.5625hz, 3.125hz, 6.25hz, 12.5hz, 25hz, 50hz, 100hz, 200hz, 500hz, 1khz, 2khz, 4khz, 8khz, 16khz, 32khz.
    The specific value depends on the current camera.
  • accel_range: The range of the accelerometer, the optional values are 2g, 4g, 8g, 16g. The specific value
    depends on the current camera.
  • enable_gyro: Enables the gyroscope.
  • gyro_rate: The frequency of the gyroscope, the optional values
    are 1.5625hz, 3.125hz, 6.25hz, 12.5hz, 25hz, 50hz, 100hz, 200hz, 500hz, 1khz, 2khz, 4khz, 8khz, 16khz, 32khz.
    The specific value depends on the current camera.
  • gyro_range: The range of the gyroscope, the optional values
    are 16dps, 31dps, 62dps, 125dps, 250dps, 500dps, 1000dps, 2000dps. The specific value depends on the
    current camera.
  • enumerate_net_device: Enables the function of enumerating network devices. True means enabled, false means disabled.
    This feature is only supported by Femto Mega and Gemini 2 XL devices. When accessing these devices through the
    network, the IP address of the device needs to be configured in advance. The enable switch needs to be set to true.
  • depth_filter_config: Configures the loading path for the depth filtering configuration file. By default, the depth
    filtering configuration file is located in the /config/depthfilter directory. Supported only on Gemini2.
  • depth_precision: The depth precision should be in the format 1mm. The default value is 1mm.
  • enable_laser: Enables the laser. The default value is true.
  • laser_on_off_mode: Laser on/off alternate mode, 0: off, 1: on-off alternate, 2: off-on alternate. The default value
    is 0.
  • device_preset: The default value is Default. Only the G330 series is supported. For more information, refer to
    the Gemini 330 documentation. Please refer to the table below to set
    the device_preset value based on your use case. The value should be one of the preset names
    listed in the table.
  • enable_decimation_filter: This filter effectively reduces the depth scene complexity. The filter runs on kernel
    sizes [2×2] to [8×8] pixels. The image size is scaled down proportionally in both dimensions to preserve the aspect
  • enable_hdr_merge: This filter is used jointly with the depth HDR function. By merging consecutive depth images of
    alternating exposure values, we can overcome challenges in acquiring depth values for under-illuminated and
    over-illuminated objects simultaneously.
  • enable_sequence_id_filter: This filter is used jointly with the depth HDR function and outputs only the sequence
    with the specified sequence ID.
  • enable_threshold_filter: This filter preserves depth values of interest and omits depth values out of scope.
  • enable_noise_removal_filter: This filter removes speckle noise in clusters and gives rise to a less-filled depth
  • enable_spatial_filter: This filter performs multiple iterations of processing as specified by the magnitude
    parameter to enhance the smoothness of depth data. It is also capable of filling small holes in depth maps.
  • enable_temporal_filter: This filter is intended to improve the depth data persistency by manipulating per-pixel
    values based on previous frames. The filter performs a single pass on the data, adjusting the depth values while also
    updating the tracking history.
  • enable_hole_filling_filter: This filter fills all holes in the depth map using the specified mode.
  • retry_on_usb3_detection_failure: If the camera is connected to a USB 2.0 port and is not detected, the system will
    attempt to reset the camera up to three times. This setting aims to prevent USB 3.0 devices from being incorrectly
    recognized as USB 2.0. It is recommended to set this parameter to false when using a USB 2.0 connection to avoid
    unnecessary resets.

IMPORTANT: Please carefully read the instructions regarding software filtering settings at this link. If you are uncertain, do not modify these settings.


Predefined presets

Preset Features Recommended use cases
Default – Best visual perception
– Overall good performance in accuracy, fill rate, tiny objects, etc.
– Generic
– Robotics
Hand – Clear hand and finger edges – Gesture recognition
High Accuracy – Depth of high confidence
– Barely noise depth values
– Lower fill rate
– Collision avoidance
– Object scanning
High Density – Higher fill rate
– More tiny objects
– May suffer from noise depth values
– Object recognition
– Pick & place
– Foreground & background animation
Medium Density – Balanced performance in fill rate and accuracy
– In comparison to Default: lower fill rate, better edge quality
– Generic and alternative to Default
Custom – User defined Preset
– Derived from Presets above, with customized modifications, e.g. a new configuration for the post-processing pipeline, modified mean intensity set point of depth AE function, etc.
– Better depth performance achieved using customized configurations in comparison to using predefined presets
– For well-established custom configurations

Choose the appropriate preset name based on your specific use case and set it as the value for the device_preset


All available service for camera control

The name of the following service already expresses its function.
However, it should be noted that the corresponding set_[ir|depth|color]*
and get[ir|depth|color]* services are only available if you set enable[ir|depth|color]
to true in the stream that corresponds to the argument of the launch file.

  • /camera/get_auto_white_balance
  • /camera/get_color_exposure
  • /camera/get_color_gain
  • /camera/get_depth_exposure
  • /camera/get_depth_gain
  • /camera/get_device_info
  • /camera/get_ir_exposure
  • /camera/get_ir_gain
  • /camera/get_ldp_status
  • /camera/get_sdk_version
  • /camera/get_white_balance
  • /camera/set_auto_white_balance
  • /camera/set_color_auto_exposure
  • /camera/set_color_exposure
  • /camera/set_color_gain
  • /camera/set_depth_auto_exposure
  • /camera/set_depth_exposure
  • /camera/set_depth_gain
  • /camera/set_fan_work_mode
  • /camera/set_floor_enable
  • /camera/set_ir_auto_exposure
  • /camera/set_ir_exposure
  • /camera/set_ir_gain
  • /camera/set_laser_enable
  • /camera/set_ldp_enable
  • /camera/set_white_balance
  • /camera/toggle_color
  • /camera/toggle_depth
  • /camera/toggle_ir


All available topics

  • /camera/color/camera_info : The color camera info.
  • /camera/color/image_raw: The color stream image.
  • /camera/depth/camera_info: The depth stream image.
  • /camera/depth/image_raw: The depth stream image
  • /camera/depth/points : The point cloud, only available when enable_point_cloud is true.
  • /camera/depth_registered/points: The colored point cloud, only available when enable_colored_point_cloud
    is true.
  • /camera/ir/camera_info: The IR camera info.
  • /camera/ir/image_raw: The IR stream image
  • /camera/accel/sample: Acceleration data stream enable_sync_output_accel_gyroturned off,enable_accelturned on
  • /camera/gyro/sample: Gyroscope data stream,enable_sync_output_accel_gyroturned off,enable_gyro`turned on
  • camera/gyro_accel/sample: Synchronized data stream of acceleration and gyroscope,enable_sync_output_accel_gyro
    turned on
  • /diagnostics: The diagnostic information of the camera, Currently, the diagnostic information only includes the temperature of the camera.


Network device enumeration

Currently, the network device enumeration function is supported only by the Femto Mega device. When accessing this
device over the network, if enumerate_net_device is set to true, the device will be automatically enumerated,
eliminating the need to configure the IP address in advance or set the enable switch to true. The specific configuration
methods are as follows:

  • enumerate_net_device: enumeration network device automatically, only supported by Femto Mega.
    if enumerate_net_device set to true, the device will be enumerated automatically,No need to set
    the net_device_ip
    and net_device_port parameters.
  • net_device_ip: The IP address of the device.
  • net_device_port: The port number of the device.



  • To get the usb_port of the camera, plug in the camera and run the following command in the terminal:
ros2 run orbbec_camera list_devices_node
  • Set the device_num parameter to the number of cameras you have.
  • Go to the OrbbecSDK_ROS2/launch/ file and change the usb_port.
  • Don’t forget to put the include tag inside the group tag.
    Otherwise, the parameter values of different cameras may become contaminated.
from launch import LaunchDescription
from launch.actions import DeclareLaunchArgument, IncludeLaunchDescription, GroupAction, ExecuteProcess
from launch.launch_description_sources import PythonLaunchDescriptionSource
from launch_ros.actions import Node
from ament_index_python.packages import get_package_share_directory
import os

def generate_launch_description():
    # Include launch files
    package_dir = get_package_share_directory('orbbec_camera')
    launch_file_dir = os.path.join(package_dir, 'launch')
    launch1_include = IncludeLaunchDescription(
            os.path.join(launch_file_dir, '')
            'camera_name': 'camera_01',
            'usb_port': '6-',  # replace your usb port here
            'device_num': '2'

    launch2_include = IncludeLaunchDescription(
            os.path.join(launch_file_dir, '')
            'camera_name': 'camera_02',
            'usb_port': '6-2.4.1',  # replace your usb port here
            'device_num': '2'

    # If you need more cameras, just add more launch_include here, and change the usb_port and device_num

    # Launch description
    ld = LaunchDescription([

    return ld

  • To launch the cameras, run the following command:
ros2 launch orbbec_camera


Use hardware decoder to decode JPEG

rockchip and Amlogic

Depends on rockchip-mpp-dev and rockchip-rga-dev, not all systems have these two packages, the names may be
different, please search by yourself.
Open CMakeLists.txt and set USE_RK_HW_DECODER to ON.

Nvidia Jetson

Depends on: jetson_multimedia_api,libyuv.
Open CMakeLists.txt and set USE_NV_HW_DECODER to ON.


Check which profiles the camera supports

ros2 run orbbec_camera list_camera_profile_mode_node


Building a Debian Package

Preparing the Environment

Before starting, install the required tools:

sudo apt install debhelper fakeroot python3-bloom

Configuring ROS Dependencies

Add the following YAML file to your system at /etc/ros/rosdep/sources.list.d/00-orbbec.yaml. Make sure to
replace focal with the codename of your Ubuntu version and humble with your ROS2 distribution name:

    focal: [ ros-humble-orbbec-camera-msgs ]

Next, create a new file /etc/ros/rosdep/sources.list.d/50-orbbec.list and add this line to specify the path to the
YAML file:

yaml file:///etc/ros/rosdep/sources.list.d/00-orbbec.yaml

Update the rosdep database to reflect these changes:

rosdep update

Building the Package

Navigate to your workspace and build the project:

cd ~/ros2_ws/
colcon build --event-handlers console_direct+ --cmake-args -DCMAKE_BUILD_TYPE=Release
. install/setup.bash
cd src/OrbbecSDK_ROS2/


Launch files

product serials launch file
gemini 330 series
femto mega
femto bolt

All launch files are essentially similar, with the primary difference being the default values of the parameters set for different models within the same series. Differences in USB standards, such as USB 2.0 versus USB 3.0, may require adjustments to these parameters. If you encounter a startup failure, please carefully review the specification manual. Pay special attention to the resolution settings in the launch file, as well as other parameters, to ensure compatibility and optimal performance.


Product support

Please refer to the OrbbecSDK supported
products: Product Support


DDS Tuning

The default DDS settings (Galactic) may not be optimal for data transmission. Different DDS settings can have varying
performance. In this example, we use CycloneDDS. For more detailed information, please refer to the
ROS DDS Tuning

● Edit cyclonedds configuration file

sudo gedit /etc/cyclonedds/config.xml


<?xml version="1.0" encoding="UTF-8"?>
<CycloneDDS xmlns="" xmlns:xsi=""
    <Domain id="any">
                <Peer address="localhost"/>

● Set the environment variables, add to .zshrc or .bashrc

export ROS_DOMAIN_ID=42 # Numbers from 0 to 232
export CYCLONEDDS_URI=file:///etc/cyclonedds/config.xml

Tip:to understand why the maximum ROS_DOMAIN_ID is 232, please
● Increase UDP receive buffer size





Frequently Asked Questions

No Picture from Multiple Cameras

  • it’s possible that the power supply is insufficient.
    To avoid this, do not connect all cameras to the same hub and use a powered hub instead.
  • It’s also possible that the resolution is too high.
    To resolve this, try lowering the resolution.

Why are there so many launch files here

  • The reason for the presence of multiple launch
    files is due to the fact that the default resolutions and image formats of different cameras vary.
    To make it easier to use, the launch files have been separated for each camera.


Other useful links



Copyright 2024 Orbbec Ltd.

Licensed under the Apache License, Version 2.0 (the “License”); you may not use this project except in compliance with
the License. You may obtain a copy of the License at

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an ”
AS IS” BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific
language governing permissions and limitations under the License.

Other names and brands may be claimed as the property of others

Stay updated

Be the first to learn about our new
products and updates.