人脸识别结构光摄像头-双目ASTRA Pro
Orbbec Astra Pro传感器在ROS（indigo和kinetic）使用说明 rgb depth同时显示这款摄像头使用uvc输入彩色信息，需要libuvc和libuvc_ros这样才能在ROS正常使用彩色功能。
请在下面网址，分别下载对应包：
1 https://github.com/ktossell
2 https://orbbec3d.com/develop/
也可以直接下载打包好的资源：
http://download.csdn.net/detail/zhangrelay/9705366
首先安装驱动，依据版本选择下载驱动中合适的版本：
https://img-blog.csdn.net/20161208113129591
依据下列文档说明进行安装：特别注意标出部分
#############Note###################
# For user with ARM based development board:
# With CPU Structure older than Cortex A17, use OpenNI-Linux-Arm-2.3 Nofilter.tar for better performance
####################################
# There are two zip files, one is for 32bit machine, the other one is for 64bit
# We choose 64bit(x64) and make the example as follows:
这里应为系统是Ubuntu 16.04 64bit PC选择对应64bit版本
# To run visual samples(e.g., SimpleViewer), you will need freeglut3 header and libaries, please install:
$ sudo apt-get install build-essential freeglut3 freeglut3-dev
#check udev version, Orbbec Driver need libudev.so.1, if can't find it, can make symbolic link from libudev.so.x.x,
#which usually locate in /lib/x86_64-linux-gnu or /lib/i386-linux-gnu
$ldconfig -p | grep libudev.so.1
$cd /lib/x86_64-linux-gnu
$sudo ln -s libudev.so.x.x.x libudev.so.1
# copy tgz file to any place you want(e.g., Home)
# unzip tgz file
拷贝到对应文件夹并解压缩：
$ tar zxvf OpenNI-Linux-x64-2.2-0118.tgz
$ cd OpenNI-Linux-x64-2.2
# run install.sh to generate OpenNIDevEnvironment, which contains OpenNI development environment
$ sudo ./install.sh
# please replug in the device for usb-register
# add environment variables
$ source OpenNIDevEnvironment
# build sample(e.g., SimpleViewer)
$ cd Samples/SimpleViewer
$ make
# run sample
# connect sensor
$ cd Bin/x64-Release
$ ./SimpleViewer
# now you should be able to see a GUI window showing the depth stream video
# If the Debian Jessie Lite is used for testing, it may require the following installation for properly start the viewer.
$ sudo apt-get install libgl1-mesa-dri
安装完毕后，进行功能测试：
~/tools/OpenNI-Linux-x64-2.3/Samples/Bin$ ls
ClosestPointViewer MultiDepthViewer org.openni.jar
EventBasedRead MultipleStreamReadorg.openni.Samples.SimpleViewer
libMWClosestPoint.soMWClosestPointApp org.openni.Samples.SimpleViewer.jar
libOpenNI2.jni.so OpenNI2 SimpleRead
libOpenNI2.so OpenNI.ini SimpleViewer
使用./SimpleViewer和./ClosestPointViewer，注意传感器特性，深度视觉0.4-8.0米，如下：
https://img-blog.csdn.net/20161208120555572
https://img-blog.csdn.net/20161208120605093
安装ROS功能包，以kinetic为例，indigo类似：
~$ sudo apt-get install ros-kinetic-astra-camera ros-kinetic-astra-launch完毕后，使用：
relaybotbox@relaybotbox-desktop:~$ roslaunch astra_launch
astra.launch device.launch pointclouds.launch
astra_tf_prefix.launchir.launch processing.launch
color.launch kinect_frames.launch
depth.launch manager.launch
relaybotbox@relaybotbox-desktop:~$ roslaunch astra_launch astra.launch
工作正常的话可以看到：
~$ roslaunch astra_launch astra.launch
... logging to /home/relaybotbox/.ros/log/26bcfad8-bcf9-11e6-bc52-00e0b4159b09/roslaunch-relaybotbox-desktop-6834.log
Checking log directory for disk usage. This may take awhile.
Press Ctrl-C to interrupt
Done checking log file disk usage. Usage is <1GB.
started roslaunch server http://relaybotbox-desktop:33177/
SUMMARY
========
PARAMETERS
* /camera/camera_nodelet_manager/num_worker_threads: 4
* /camera/depth_rectify_depth/interpolation: 0
* /camera/driver/auto_exposure: True
* /camera/driver/auto_white_balance: True
* /camera/driver/color_depth_synchronization: False
* /camera/driver/depth_camera_info_url:
* /camera/driver/depth_frame_id: camera_depth_opti...
* /camera/driver/depth_registration: False
* /camera/driver/device_id: #1
* /camera/driver/rgb_camera_info_url:
* /camera/driver/rgb_frame_id: camera_rgb_optica...
* /rosdistro: kinetic
* /rosversion: 1.12.6
NODES
/camera/
camera_nodelet_manager (nodelet/nodelet)
depth_metric (nodelet/nodelet)
depth_metric_rect (nodelet/nodelet)
depth_points (nodelet/nodelet)
depth_rectify_depth (nodelet/nodelet)
depth_registered_sw_metric_rect (nodelet/nodelet)
driver (nodelet/nodelet)
points_xyzrgb_sw_registered (nodelet/nodelet)
register_depth_rgb (nodelet/nodelet)
rgb_rectify_color (nodelet/nodelet)
/
camera_base_link (tf2_ros/static_transform_publisher)
camera_base_link1 (tf2_ros/static_transform_publisher)
camera_base_link2 (tf2_ros/static_transform_publisher)
camera_base_link3 (tf2_ros/static_transform_publisher)
auto-starting new master
process: started with pid
ROS_MASTER_URI=http://localhost:11311
setting /run_id to 26bcfad8-bcf9-11e6-bc52-00e0b4159b09
process: started with pid
started core service
process: started with pid
process: started with pid
process: started with pid
process: started with pid
process: started with pid
process: started with pid
process: started with pid
process: started with pid
process: started with pid
process: started with pid
process: started with pid
process: started with pid
process: started with pid
[ INFO] : Initializing nodelet with 4 worker threads.
[ INFO] : Device "2bc5/0403@2/5" found.
Warning: USB events thread - failed to set priority. This might cause loss of data...
使用rqt_image_view：
https://img-blog.csdn.net/20161208120748619
https://img-blog.csdn.net/20161208120800619
发现rgb并没有正确显示？？需要UVC支持。
编译libuvc过程如下：注意红色部分
`libuvc` is a cross-platform library for USB video devices, built atop `libusb`.
It enables fine-grained control over USB video devices exporting the standard USB Video Class
(UVC) interface, enabling developers to write drivers for previously unsupported devices,
or just access UVC devices in a generic fashion.
## Getting and Building libuvc
Prerequisites: You will need `libusb` and (http://www.cmake.org/) installed.
To build, you can just run these shell commands:
git clone https://github.com/ktossell/libuvc
cd libuvc
mkdir build
cd build
cmake ..
make && sudo make install
and you're set! If you want to change the build configuration, you can edit `CMakeCache.txt`
in the build directory, or use a CMake GUI to make the desired changes.
## Developing with libuvc
The documentation for `libuvc` can currently be found at https://int80k.com/libuvc/doc/.
然后编译libuvc_ros：注意红色部分
`libuvc_camera` is a ROS driver that supports webcams and other UVC-standards-compliant video devices.
It's a cross-platform replacement for `uvc_camera`, a Linux-only webcam driver.
Documentation is available on the ROS wiki: (http://wiki.ros.org/libuvc_camera).
使用catkin_make，如果报错，注意libusb.h的位置。使用locate libusb.h，并将其放到合适位置：
~$ sudo cp /usr/include/libusb-1.0/libusb.h/usr/local/include/libuvc/
修改下libuvc.h中include中的路径。
在启动roscore后，使用：
~$ rosrun libuvc_camera camera_node
查看rgb。
~$ rosrun libuvc_camera camera_node
[ INFO] : Opening camera with vendor=0x0, product=0x0, serial="", index=0
: Permission denied opening /dev/bus/usb/002/004出现报错，修改端口权限，重试。
: Permission denied opening /dev/bus/usb/002/004
relaybotbox@relaybotbox-desktop:~$ sudo chmod 777 /dev/bus/usb/002/004
relaybotbox 的密码：
relaybotbox@relaybotbox-desktop:~$ rosrun libuvc_camera camera_node
[ INFO] : Opening camera with vendor=0x0, product=0x0, serial="", index=0
unsupported descriptor subtype: 13
unsupported descriptor subtype: 13
[ WARN] : Unable to set scanning_mode to 0
[ WARN] : Unable to set auto_exposure to 8
[ WARN] : Unable to set auto_exposure_priority to 0
[ WARN] : Unable to set exposure_absolute to 1
[ WARN] : Unable to set auto_focus to 1
[ WARN] : Unable to set focus_absolute to 0
[ WARN] : Unable to set gain to 0
[ WARN] : Unable to set iris_absolute to 0
[ WARN] : Unable to set pantilt to 0, 0
[ INFO] : using default calibration URL
[ INFO] : camera calibration URL: file:///home/relaybotbox/.ros/camera_info/camera.yaml
[ INFO] : Unable to open camera calibration file
[ WARN] : Camera calibration file /home/relaybotbox/.ros/camera_info/camera.yaml not found.
使用 rqt_image_view ，查看rgb如下：
https://img-blog.csdn.net/20161208120910910
同时使用也是可以的。
https://img-blog.csdn.net/20161208120924770
https://img-blog.csdn.net/20161208120939714
安装配置完毕，可以开始使用这款视觉传感器了。
---------------------
作者：zhangrelay
来源：CSDN
原文：https://blog.csdn.net/zhangrelay/article/details/53515859
版权声明：本文为博主原创文章，转载请附上博文链接！
页:
[1]