Thank you for your concern.
Since i don't have time to update this repo, also intel has release their own official version.
So if you want to support all the types of RealSense camera please visit
https://github.com/intel-ros/realsense
https://github.com/IntelRealSense/librealsense
This repo will change to research.
THX
DD
use realsense camera in ROS
- [2015/11/26] Add print center Z value and realsense temperature at the same time for
F200
.
Can use to calibration Z value with realsense's temperature(Thx to @teknotus).
0) sudo apt-get install libusb-1.0-0-dev
1) sudo cp config/80-realsense-usb.rules /etc/udev/rules.d
2) reboot to reload the rule
3) rebuild realsense_camera
4) roslaunch realsense_camera realsense_rviz.launch debug_depth_unit:=true
If everything is OK, in terminal you will see
getRealsenseUSBHandle OK!
average center z value = 0.388359 temp = 31 depth_unit = 33.099998
-
Modify default image topic name like openni.
-
Add R200 support, But still in test, Because i don't have the R200 device now.
-
Intel maybe release official ROS-Realsense package.
visit http://roscon.ros.org/2015/ then search realsense
-
[2015/04/10] Added infrared stream for realsense camera
If you want IR stream, please visit http://solsticlipse.com/2015/03/31/intel-real-sense-3d-on-linux-macos.html https://github.com/teknotus/depthview/tree/kernelpatchfmt
This package do not contains uvmap data(use to register RGB data)
So it is only publish point cloud without RGB
If you want RGB data for point cloud
Please visit realsense_camera_tools
You can copy the uvmap files by realsense's serial number to data/uvmap/
realsense_camera
├── data
│ └── uvmap
│ ├── 039140070906
│ │ ├── uvmap_0000.bin
│ │ ├── uvmap_0001.bin
│ │ ├── ...
│ │ ├── ...
│ │ ├── uvmap_2047.bin
039140070906 is realsense's serial number
sudo apt-get install libudev-dev libv4l-dev libusb-1.0-0-dev
To make sure the dynamic reconfigure works, Please visit
https://github.com/teknotus/depthview/tree/udev
- launch realsense camera and rviz
roslaunch realsense_camera realsense_rviz.launch
- launch realsense camera only
roslaunch realsense_camera realsense_camera.launch
default
sensor_msgs::PointCloud2
/camera/depth/points point cloud without RGB
/camera/depth_registered/points point cloud with RGB
sensor_msgs::Image
/camera/rgb/image_raw raw image for RGB sensor
/camera/depth/image_raw raw image for depth sensor
/camera/ir/image_raw raw image for infrared sensor
you can custom topic in file realsense_camera.launch
<arg name="topic_depth_points_id" default="$(arg camera)/depth/points" />
<arg name="topic_depth_registered_points_id" default="$(arg camera)/depth_registered/points" />
<arg name="topic_image_rgb_raw_id" default="$(arg camera)/rgb/image_raw" />
<arg name="topic_image_depth_raw_id" default="$(arg camera)/depth/image_raw" />
<arg name="topic_image_infrared_raw_id" default="$(arg camera)/ir/image_raw" />
To specify a yaml file with intrinsic calibration parameters for the RGB & IR camera (for example generated using this tool), modify the value in the realsense_(r200_)camera.launch to your calibration file (eg. package://package_name/path_to_yaml_file/calibration_file.yaml )
<!-- camera calibration file url -->
<arg name="rgb_camera_info_url" default="" />
<arg name="ir_camera_info_url" default="" />
Make sure the image width and height specified in the calibration file is the same as that specified in launch file.
The calibration info will be published under the topic /camera/rgb/camera_info
/camera/depth/camera_info
/camera/ir/camera_info
- If have "select timeout" issue
Stop realsense_camera node
Re-plug realsense device then retry
- RGB to depth misalignment
- add use TF between RGB & depth to register rgb data
- add usb reset
- add read properties from device
- add RGB to depth camera calibration
terminate called after throwing an instance of 'boost::exception_detail::clone_impl<boost::exception_detail::error_info_injector<boost::lock_error> >' what(): boost: mutex lock failed in pthread_mutex_lock: Invalid argument
//fill point cloud data time: [0.035777 s] <---- need optimize