08-Server Side

In the previous lesson, we discussed how the client requests services and how the server provides services. In this section, we focus on how to implement a service server.

---

8.1 C++ Language Implementation

8.1.1 Implementation Steps

1. Initialize the ROS node\
2. Create a server instance\
3. Loop while waiting for service requests and enter the callback function\
4. Process the service logic in the callback and return a response

---

8.1.2 Create the Server Program

Switch to:

~/catkin_ws/src/learning_server/src

Create the following file:

turtle_vel_command_server.cpp

Paste the following code:

#include <ros/ros.h>
#include <geometry_msgs/Twist.h>
#include <std_srvs/Trigger.h>

ros::Publisher turtle_vel_pub;
bool pubvel = false;

bool pubvelCallback(std_srvs::Trigger::Request &req,
                    std_srvs::Trigger::Response &res)
{
    pubvel = !pubvel;

    ROS_INFO("Do you want to publish the vel?: [%s]", pubvel == true ? "Yes" : "No");

    res.success = true;
    res.message = "The status is changed!";

    return true;
}

int main(int argc, char **argv)
{
    ros::init(argc, argv, "turtle_vel_command_server");
    ros::NodeHandle n;

    ros::ServiceServer command_service =
        n.advertiseService("/turtle_vel_command", pubvelCallback);

    turtle_vel_pub =
        n.advertise<geometry_msgs::Twist>("/turtle1/cmd_vel", 8);

    ros::Rate loop_rate(10);

    while (ros::ok())
    {
        ros::spinOnce();

        if (pubvel)
        {
            geometry_msgs::Twist vel_msg;
            vel_msg.linear.x = 0.6;
            vel_msg.angular.z = 0.8;
            turtle_vel_pub.publish(vel_msg);
        }

        loop_rate.sleep();
    }

    return 0;
}

---

Process Flow Diagram

---

CMakeLists.txt Configuration

Add the following under the build section:

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

---

Compile the Workspace

cd ~/catkin_ws
catkin_make
source devel/setup.bash

---

Run the Program

roscore
rosrun turtlesim turtlesim_node
rosrun learning_server turtle_vel_command_server

---

Runtime Effect

---

Program Description

After starting the turtlesim node, you can list available services:

rosservice list

After running the server, you will see the /turtle_vel_command service appear:

Calling the service toggles the movement of the turtle. If the service is called once, the turtle starts moving in a circular motion. If called again, the turtle stops. This is achieved by toggling the pubvel flag inside the service callback.

---

8.2 Python Language Implementation

8.2.1 Create the Python Server Script

Switch to:

~/catkin_ws/src/learning_server/scripts

Create the following file:

turtle_vel_command_server.py

Paste the following code:

#!/usr/bin/env python
# -*- coding: utf-8 -*-

import rospy
import thread, time
from geometry_msgs.msg import Twist
from std_srvs.srv import Trigger, TriggerResponse

pubvel = False
turtle_vel_pub = rospy.Publisher('/turtle1/cmd_vel', Twist, queue_size=8)

def pubvel_thread():
    while True:
        if pubvel:
            vel_msg = Twist()
            vel_msg.linear.x = 0.6
            vel_msg.angular.z = 0.8
            turtle_vel_pub.publish(vel_msg)
        time.sleep(0.1)

def pubvelCallback(req):
    global pubvel
    pubvel = bool(1 - pubvel)

    rospy.loginfo("Do you want to publish the vel? [%s]", pubvel)

    return TriggerResponse(1, "Change state!")

def turtle_pubvel_command_server():
    rospy.init_node('turtle_vel_command_server')

    s = rospy.Service('/turtle_vel_command', Trigger, pubvelCallback)

    print("Ready to receive turtle_pub_vel_command.")

    thread.start_new_thread(pubvel_thread, ())
    rospy.spin()

if __name__ == "__main__":
    turtle_pubvel_command_server()

---

Python Server Flow Diagram