AR Drone系列之：使用ROS catkin创建package并使用cv_bridge实现对ar drone摄像头数据的处理

1 开发环境

Ubuntu 12.04

ROS Hydro

2 前提

可參考这篇blog：http://blog.csdn.net/yake827/article/details/44564057

blog：http://blog.csdn.net/celesius/article/details/39188119

已安装adrone_autonomy package 而且能够执行

https://github.com/AutonomyLab/ardrone_autonomy

文档：http://ardrone-autonomy.readthedocs.org

已通过catkin创建一个package （方法见上一篇文章）这里我创建的名称为droneTest

3 欲实现效果

获取ar drone的摄像头实时图像而且能够进行处理

4 參考网页

http://answers.ros.org/question/79306/help-with-streaming-ardrone-camera-images-to-opencv/

http://wiki.ros.org/cv_bridge/Tutorials/UsingCvBridgeToConvertBetweenROSImagesAndOpenCVImages

http://wiki.ros.org/vision_opencv

5 详细实现Step-by-Step

Step 1：在~/catkin_ws/src/droneTest/src/ 中创建一个新的文件这里命名为droneTest.cpp

Step 2: 编辑droneTest.cpp文件，代码例如以下：

#include <ros/ros.h>

#include <image_transport/image_transport.h>

#include <sensor_msgs/image_encodings.h>

#include <cv_bridge/cv_bridge.h>

#include <opencv2/highgui/highgui.hpp>

#include <opencv2/imgproc/imgproc.hpp>

using namespace std;

using namespace cv;

static const char WINDOW[]="RGB Image";

static const char WINDOW2[]="Gray Image";

void process(const sensor_msgs::ImageConstPtr& cam_image){

cv_bridge::CvImagePtr cv_ptr;

try

{

  cv_ptr = cv_bridge::toCvCopy(cam_image,sensor_msgs::image_encodings::BGR8);

}

catch (cv_bridge::Exception& e)

{

  ROS_ERROR("cv_bridge exception:%s",e.what());

  return;

}

Mat img_rgb = cv_ptr->image;

Mat img_gray;

cvtColor(img_rgb,img_gray,CV_RGB2GRAY);

imshow(WINDOW,img_rgb);

imshow(WINDOW2,img_gray);

cvWaitKey(1);

}

int main(int argc, char **argv){

ros::init(argc,argv,"droneTest");

ros::NodeHandle n;

image_transport::ImageTransport it(n);

image_transport::Subscriber image_sub = it.subscribe("/ardrone/image_raw",1,process);

cv::namedWindow(WINDOW);

cv::namedWindow(WINDOW2);

ros::spin();

return 0;

}

这里使用cv_bridge的toCvCopy来实现格式转换。很easy

Step 3：编辑CMakeLists.txt

主要目的是加入依赖和加入opencv库

cmake_minimum_required(VERSION 2.8.3)

project(droneTest)

find_package(catkin REQUIRED COMPONENTS

  roscpp

  std_msgs

  sensor_msgs

  cv_bridge

  image_transport

)

catkin_package()

find_package(OpenCV)

include_directories(

  ${OpenCV_INCLUDE_DIRS}

)

include_directories(include ${catkin_INCLUDE_DIRS})

add_executable(droneTest src/droneTest.cpp)

target_link_libraries(droneTest ${catkin_LIBRARIES} ${OpenCV_LIBRARIES})

add_dependencies(droneTest droneTest_generate_messages_cpp)

Step 4:编译

编译catkin。在terminal中输入：

cd ~/catkin_ws

catkin_make

这里说明一下就是package.xml这个文件改不改不影响，我发现甚至把里面的dependency都删掉也能够make。

接下来是执行

这里我为了执行方便一般把package拷贝到~/workshop下

然后把~/catkin_ws/devel/lib/droneTest 拷贝到~/workshop/droneTest下。这里我的ROS_PACKAGE_PATH 包括~/workshop

我在bashrc中有加入例如以下代码：

source /opt/ros/hydro/setup.bash

export ROS_PACKAGE_PATH=$ROS_PACKAGE_PATH:~/workshop

Step 6：执行

1打开一个terminal执行roscore

2 连接ar drone

3 再打开一个terminal执行rosrun ardrone_autonomy ardrone_driver

4 再打开一个terminal执行rosrun droneTest droneTest

ok了