basic_ros_sub.md

First ROS Subscriber

Video Tutorial at https://youtu.be/iueRUQCvJXw

In this tutorial I will be showing you how to write a ROS subscriber. A ROS subscriber allows you to use data being published on a ROS topic in your own C++ ROS node.

Create a New C++ File

in iq_gnc/src create a new file and call it sub.cpp

Add C++ File to CMakeLists

Next open up the CMakeLists.txt in iq_gnc and add the following lines to the bottom of the file

add_executable(sub src/sub.cpp)
target_link_libraries(sub ${catkin_LIBRARIES})

these will allow catkin to build our program when we next run catkin build

Add Includes and Main Function

in every ros program we make, we will need to add the ros include as follows

#include <ros/ros.h>

then below, we will add our main function.

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

	//rest of code will go here 

	return 0;
}

Initialize ROS

	ros::init(argc, argv, "detection_sub");
	ros::NodeHandle n;

Declare our Subscriber

subscribers are declared in the form

	ros::Subscriber sub = <nodehandle>.subscribe("<topic>", <# of msg buffered>, <name of callback function>);

We want to subscribe to the detection data coming from darknet, so lets see which topics might have the data we need. run

roslaunch darknet_ros darknet_ros.launch 

then

rostopic list 

we see the following topics

/darknet_ros/bounding_boxes
/darknet_ros/check_for_objects/cancel
/darknet_ros/check_for_objects/feedback
/darknet_ros/check_for_objects/goal
/darknet_ros/check_for_objects/result
/darknet_ros/check_for_objects/status
/darknet_ros/detection_image
/darknet_ros/found_object

Lets subscribe to /darknet_ros/bounding_boxes, but fist we need to see what type of ROS message this is. run

rostopic list -v /darknet_ros/bounding_boxes

we see

Published topics:
 * /darknet_ros/bounding_boxes [darknet_ros_msgs/BoundingBoxes] 1 publisher

Subscribed topics:

this means that the topic is publishing data in the form of darknet_ros_msgs/BoundingBoxes

Now we should have enough info to declare our subscriber. write

	ros::Subscriber sub = n.subscribe("/darknet_ros/bounding_boxes", 1, detection_cb);

then add ros::spin(); this way the subscriber continues to get the latest data as it becomes available.

ros::spin();

we will also need to include the ros message to our include section. At the top of the program add

#include <darknet_ros_msgs/BoundingBoxes.h>

Write our Callback Function

we are going to call our call back function detection_cb. go ahead and add the following code

void detection_cb(const darknet_ros_msgs::BoundingBoxes::ConstPtr& msg)
{	
	//rest of callback function code
}

now we want to get the name of the object we are seeing out of the message, so lets learn more about the structure of this message. run the following

rosmsg show darknet_ros_msgs/BoundingBoxes

you will see

std_msgs/Header header
  uint32 seq
  time stamp
  string frame_id
std_msgs/Header image_header
  uint32 seq
  time stamp
  string frame_id
darknet_ros_msgs/BoundingBox[] bounding_boxes
  string Class            <---- the information we want
  float64 probability
  int64 xmin
  int64 ymin
  int64 xmax
  int64 ymax

we can access the above information by dereferencing the ROS message as follows msg->bounding_boxes[i].Class. add the following code to the callback function to print out all the objects detected.

	for( int i=0; i < msg->bounding_boxes.size(); i++)
	{
		ROS_INFO("%s detected", msg->bounding_boxes[i].Class.c_str());	
	}	

Build and Run Program

build

catkin build 

then run our simulation and takeoff the drone

we can run our new program by running

rosrun iq_gnc sub 

we should see the objects seen by the drone being printed in the console

---

References

http://wiki.ros.org/ROS/Tutorials/WritingPublisherSubscriber%28c%2B%2B%29