PX4 GAZEBO无人机添加摄像头并进行图像识别
在之前完成了ROS的安装和PX4的安装,并可以通过roslaunch启动软件仿真。接下来为无人及添加相机,并将图像用python函数读取,用于后续操作
-
添加相机,参考:ROS PX4添加相机
-
读取摄像头图像
rqt_image_view
- 高级用法,自己添加model,参考:添加向下的相机
但是并没有成功,博主提供了github的链接,但不知道从何处下手。 - 添加激光雷达、前摄像头,向下摄像头,成功了。参考:添加激光雷达和摄像头
总结操作方法:
1. 修改mavros_posix_sitl.launch文件
添加内容
<arg name="my_model" default="iris_fusion"/>
修改内容,注意需要修改udp的端口,否则无法获取到传感器数据
# <arg name="sdf" default="$(find mavlink_sitl_gazebo)/models/$(arg my_model)/$(arg my_model).sdf"/>
<arg name="sdf" default="$(find mavlink_sitl_gazebo)/models/$(arg vehicle)/$(arg vehicle).sdf"/>
# <arg name="fcu_url" default="udp://:14540@localhost:14557"/>
<arg name="fcu_url" default="udp://:24540@localhost:14557"/>
2. models文件夹下增加iris_fusion
文件夹:PX4_Firmware/Tools/sitl_gazebo/models/
- 添加文件夹iris_fusion
- 添加文件iris_fusion.sdf
<?xml version='1.0'?>
<sdf version='1.5'>
<model name='iris_fpv_cam'>
<include>
<uri>model://iris</uri>
</include>
<!-- Built-in depth camera -->
<include>
<uri>model://fpv_cam_self</uri>
<pose>0.1 0 0 0 0.57 0</pose>
</include>
<joint name="fpv_cam_self_joint" type="fixed">
<child>fpv_cam_self::link</child>
<parent>iris::base_link</parent>
<axis>
<xyz>0 0 1</xyz>
<limit>
<upper>0</upper>
<lower>0</lower>
</limit>
</axis>
</joint>
<include>
<uri>model://fpv_cam</uri>
<pose>0 0 0 0 1.57 0</pose>
</include>
<joint name="fpv_cam_joint" type="fixed">
<child>fpv_cam::link</child>
<parent>iris::base_link</parent>
<axis>
<xyz>0 0 1</xyz>
<limit>
<upper>0</upper>
<lower>0</lower>
</limit>
</axis>
</joint>
</model>
</sdf>
- 添加文件model.config
<?xml version="1.0"?>
<model>
<name>3DR Iris with FPV camera</name>
<version>1.0</version>
<sdf version='1.4'>iris_fusion.sdf</sdf>
<author>
<name>Amy Wagoner</name>
<email>arwagoner@gmail.com</email>
</author>
<description>
This is a model of the 3DR Iris Quadrotor with an FPV camera. The original model has been created by
Thomas Gubler and is maintained by Lorenz Meier.
</description>
</model>
3. models文件夹下增加fpv_cam_self
- 添加文件夹fpv_cam_self
- 添加文件fpv_cam_self.sdf
<?xml version="1.0" ?>
<sdf version="1.5">
<model name="fpv_cam_self">
<pose>-2 2.5 1 0 0 0</pose>
<link name="link">
<inertial>
<mass>0.01</mass>
<inertia>
<ixx>4.16666666667e-06</ixx>
<iyy>5.20833333333e-07</iyy>
<izz>3.85416666667e-06</izz>
</inertia>
</inertial>
<collision name="collision">
<geometry>
<box>
<size>0.023000 0.076000 0.032000</size>
</box>
</geometry>
</collision>
<visual name="visual">
<geometry>
<box>
<size>0.023000 0.076000 0.032000</size>
</box>
</geometry>
</visual>
<sensor type="depth" name="camera">
<always_on>true</always_on>
<update_rate>30.0</update_rate>
<camera>
<horizontal_fov>1.3962634</horizontal_fov>
<image>
<format>B8G8R8</format>
<width>640</width>
<height>480</height>
</image>
<clip>
<near>0.4</near>
<far>16.0</far>
</clip>
</camera>
<plugin name="camera_plugin" filename="libgazebo_ros_openni_kinect.so">
<baseline>0.2</baseline>
<alwaysOn>true</alwaysOn>
<!-- Keep this zero, update_rate in the parent <sensor> tag
will control the frame rate. -->
<updateRate>0.0</updateRate>
<cameraName>camera_ir</cameraName>
<imageTopicName>/camera/image_raw</imageTopicName>
<cameraInfoTopicName>/camera/camera_info</cameraInfoTopicName>
<depthImageTopicName>/camera/depth/image_raw</depthImageTopicName>
<depthImageCameraInfoTopicName>/camera/depth/camera_info</depthImageCameraInfoTopicName>
<pointCloudTopicName>/camera/depth/points</pointCloudTopicName>
<frameName>camera_link</frameName>
<pointCloudCutoff>0.5</pointCloudCutoff>
<pointCloudCutoffMax>8.0</pointCloudCutoffMax>
<distortionK1>0</distortionK1>
<distortionK2>0</distortionK2>
<distortionK3>0</distortionK3>
<distortionT1>0</distortionT1>
<distortionT2>0</distortionT2>
<CxPrime>0</CxPrime>
<Cx>0</Cx>
<Cy>0</Cy>
<focalLength>180</focalLength>
<hackBaseline>0</hackBaseline>
</plugin>
</sensor>
<gravity>0</gravity>
</link>
</model>
</sdf>
- 添加文件model.config
<?xml version="1.0"?>
<model>
<name>fpv_cam_self</name>
<version>1.0</version>
<sdf version="1.5">fpv_cam_self.sdf</sdf>
<description>
Work in progress.
kinetic 3D camera
</description>
</model>
4. 测试
# 运行
roslaunch px4 mavros_posix_sitl.launch
# 查看图像
rqt_image_view