ros学习之topic实例5

接上次的实例4,这次就用类来实现，接收，处理，并重新发布

先创建包

 catkin_create_pkg pcl_process_class roscpp std_msgs

在对应的src下面写入 pcl_process_class.cpp 代码如下

>>#include <ros/ros.h>
>>#include <pcl/point_types.h>
>>#include <pcl_conversions/pcl_conversions.h>
>>#include <sensor_msgs/PointCloud2.h>
   
  class pcl_sub_process_pub{
      private:
>>        ros::NodeHandle nh;
>>        ros::Subscriber subCloud;
>>        ros::Publisher pubCloud;
>>        sensor_msgs::PointCloud2 process_msg;
>>        pcl::PointCloud<pcl::PointXYZRGB> process_pcl;
   
      public:
          pcl_sub_process_pub():
              nh("~"){
>>            subCloud = nh.subscribe<sensor_msgs::PointCloud2>("/color_cloud", 1, &pcl_sub_process_pub::getcloud, this);
>>            pubCloud = nh.advertise<sensor_msgs::PointCloud2>("/processed_cloud_red", 1000);
   
          }
   
          //callback
>>        void getcloud(const sensor_msgs::PointCloud2ConstPtr& laserScanMsg){
  
>>            pcl::PointCloud<pcl::PointXYZRGB>::Ptr process_pcl_ptr(new pcl::PointCloud<pcl::PointXYZRGB>);
   
              //把msg给到pcl
>>            pcl::fromROSMsg(*laserScanMsg, *process_pcl_ptr);
>>            process_pcl = *process_pcl_ptr;
   
              //处理点云再发布出去.
              for (int i=0; i<process_pcl.points.size(); i++){
                  process_pcl.points[i].r = 255;
                  process_pcl.points[i].g = 0;
                  process_pcl.points[i].b = 0;
              }
              //重新转成msg发布出去
>>            pcl::toROSMsg(process_pcl, process_msg);
              pubCloud.publish(process_msg);
          }
                                                                                                                                                                                                                                                              
          ~pcl_sub_process_pub(){}
  };
   
  int main(int argc, char** argv){
>>    ros::init(argc, argv, "change_color");
      pcl_sub_process_pub ps;
>>    ros::spin();
      return 0;
  }
                                                        

CMake代如下

cmake_minimum_required(VERSION 2.6 FATAL_ERROR)
project(pcl_process_class)
 
set(PACKAGE_DEPENDENCIES
      roscpp
        sensor_msgs
          pcl_ros
            pcl_conversions
              std_srvs
                message_generation                                                                                                                                                                                                                            
                  std_msgs
                  )
 
find_package(PCL 1.3 REQUIRED COMPONENTS common io roscpp rospy std_msgs)
find_package(catkin REQUIRED COMPONENTS ${PACKAGE_DEPENDENCIES})
 
catkin_package(
#  INCLUDE_DIRS include
#  LIBRARIES publish_subscribe_demo
  CATKIN_DEPENDS  roscpp std_msgs
#  DEPENDS system_lib
)
 
include_directories(include ${catkin_INCLUDE_DIRS} ${PCL_INCLUDE_DIRS})
link_directories(${PCL_LIBRARY_DIRS})
add_definitions(${PCL_DEFINITIONS})
add_executable(pcl_process_class src/pcl_process_class.cpp)
target_link_libraries(pcl_process_class ${PCL_LIBRARIES}  ${catkin_LIBRARIES} )

编译如下

catkin_make -DCATKIN_WHITELIST_PACKAGES="pcl_process_class"

开多个tmux窗口分别运行

roscore
rosrun pub_pcl pub_pcl
rosrun pcl_process_class pcl_process_class
rosrun pcl_rpocess pcl_process
rosrun rviz rviz

就可以看到原始点云，以及分别改成绿色和红色的点云.