Mastering ROS for Robotics Programming - Second Edition

A typical structure of an ROS package is shown here:

Figure 3: Structure of a typical C++ ROS package

We can discuss the use of each folder as follows:

We need to know some commands to create, modify, and work with the ROS packages. Here are some of the commands used to work with ROS packages:

To work with packages, ROS provides a bash-like command called rosbash (http://wiki.ros.org/rosbash), which can be used to navigate and manipulate the ROS package. Here are some of the rosbash commands:

The definition of package.xml of a typical package is shown in the following screenshot:

Figure 4: Structure of package.xml

The package.xml file consists of the package name, version of the package, the package description, author details, package build dependencies, and runtime dependencies. The <build_depend></build_depend> tag includes the packages that are necessary to build the source code of the package. The packages inside the <run_depend></run_depend> tags are necessary during runtime of the package node.

Metapackages are specialized packages in ROS that only contain one file, that is, a package.xml file. They don't contain folders and files like a normal package.

Metapackages simply group a set of multiple packages as a single logical package. In the package.xml file, the metapackage contains an export tag, as shown here:

  <export> 
    <metapackage/> 
  </export>  

Also, in metapackages, there are no <buildtool_depend> dependencies for catkin; there are only <run_depend> dependencies, which are the packages grouped in the metapackage.

The ROS navigation stack is a good example of metapackages. If ROS and its navigation package are installed, we can try the following command, by switching to the navigation metapackage folder:

$ roscd navigation

Open package.xml using your favorite text editor (gedit in the following case):

$ gedit package.xml

This is a lengthy file; here is a stripped-down version of it:

Figure 5: Structure of meta-package package.xml

The ROS nodes can write or read data that has a different type. The types of data are described using a simplified message description language, also called ROS messages. These datatype descriptions can be used to generate source code for the appropriate message type in different target languages.

The data type description of ROS messages is stored in .msg files in the msg subdirectory of a ROS package. Even though the ROS framework provides a large set of robotic-specific messages already implemented, developers can define their own message type inside their nodes.

The message definition can consist of two types: fields and constants. The field is split into field types and field names. The field type is the data type of the transmitting message and field name is the name of it. The constants define a constant value in the message file.

Here is an example of message definitions:

int32 number 
string name 
float32 speed 

Here, the first part is the field type and the second is the field name. The field type is the data type and the field name can be used to access the value from the message. For example, we can use msg.number for accessing the value of the number from the message.

Here is a table showing some of the built-in field types that we can use in our message:

Other kinds of messages are designed to cover a specific application necessity, such as exchanging common geometrical (geometry_msgs) or sensor (sensor_msgs) information. A special type of ROS message is called a message header. Headers can carry information, such as time, frame of reference or frame_id, and sequence number. Using headers, we will get numbered messages and more clarity in who is sending the current message. The header information is mainly used to send data such as robot joint transforms (TF). Here is an example of the message header:

uint32 seq 
time stamp 
string frame_id 

The rosmsg command tool can be used to inspect the message header and the field types. The following command helps to view the message header of a particular message:

$ rosmsg show std_msgs/Header

This will give you an output like the preceding example message header. We will look at the rosmsg command and how to work with custom message definitions further in the upcoming sections.

The ROS services are a type request/response communication between ROS nodes. One node will send a request and wait until it gets a response from the other. The request/response communication is also using the ROS message description.

Similar to the message definitions using the ".msg" file, we have to define the service definition in another file called ".srv", which has to be kept inside the srv subdirectory of the package. Similar to the message definition, a service description language is used to define the ROS service types.

An example service description format is as follows:

#Request message type 
string str 
--- 
#Response message type 
string str 

The first section is the message type of the request that is separated by --- and in the next section is the message type of the response. In these examples, both Request and Response are strings.

In the upcoming sections, we will look at how to work with ROS services.

ROS nodes are a process that perform computation using ROS client libraries such as roscpp and rospy. One node can communicate with other nodes using ROS Topics, Services, and Parameters.

A robot might contain many nodes; for example, one node processes camera images, one node handles serial data from the robot, one node can be used to compute odometry, and so on.

Using nodes can make the system fault tolerant. Even if a node crashes, an entire robot system can still work. Nodes also reduce the complexity and increase debug-ability compared to monolithic code because each node is handling only a single function.

All running nodes should have a name assigned to identify them from the rest of the system. For example, /camera_node could be a name of a node that is broadcasting camera images.

There is a rosbash tool to introspect ROS nodes. The rosnode command can be used to get information about a ROS node. Here are the usages of rosnode:

We will look at example nodes using the roscpp client and will discuss the working of ROS nodes that use functionalities such ROS Topics, Service, Messages, and actionlib.

ROS nodes communicate with each other by publishing messages to a topic. As we discussed earlier, messages are a simple data structure containing field types. The ROS message supports standard primitive datatypes and arrays of primitive types.

Nodes can also exchange information using service calls. Services are also messages. The service message definitions are defined inside the srv file.

We can access the message definition using the following method. For example, to access std_msgs/msg/String.msg, we can use std_msgs/String. If we are using the roscpp client, we have to include std_msgs/String.h for the string message definition.

In addition to message data type, ROS uses an MD5 checksum comparison to confirm whether the publisher and subscriber exchange the same message data types.

ROS has inbuilt tools called rosmsg to get information about ROS messages. Here are some parameters used along with rosmsg:

ROS topics are named buses in which ROS nodes exchange messages. Topics can anonymously publish and subscribe, which means that the production of messages is decoupled from the consumption. The ROS nodes are not interested in knowing which node is publishing the topic or subscribing topics; they only look for the topic name and whether the message types of the publisher and subscriber are matching.

The communication using topics are unidirectional. If we want to implement a request/response, such as communication, we have to switch to ROS services.

The ROS nodes communicate with topics using TCP/IP-based transport known as TCPROS. This method is the default transport method used in ROS. Another type of communication is UDPROS, which has low-latency, loose transport, and is only suited for teleoperations.

The ROS topic tool can be used to get information about ROS topics. Here is the syntax of this command:

When we need a request/response kind of communication in ROS, we have to use the ROS services. ROS topics can't implement natively such kind of communication because it is unidirectional. The ROS services are mainly used in a distributed system.

The ROS services are defined using a pair of messages. We have to define a request datatype and a response datatype in a srv file. The srv files are kept in a srv folder inside a package.

In ROS services, one node acts as a ROS server in which the service client can request the service from the server. If the server completes the service routine, it will send the results to the service client. For example, consider a node able to provide the sum of two numbers received in input, implementing this functionality through a ROS service. The other nodes of our system might request the sum of two numbers via this service. Differently, topics are used to stream continuous data flow.

The ROS service definition can be accessed by the following method; for example, if my_package/srv/Image.srv can be accessed by my_package/Image.

In ROS services also, there is an MD5 checksum that checks in the nodes. If the sum is equal, then only the server responds to the client.

There are two ROS tools to get information about the ROS service. The first tool is rossrv, which is similar to rosmsg, and is used to get information about service types. The next command is rosservice, which is used to list and query about the running ROS services.

The following explain how to use the rosservice tool to get information about the running services:

A bag file in ROS is for storing ROS message data from topics and services. The .bag extension is used to represent a bag file.

Bag files are created using the rosbag command, which will subscribe one or more topics and store the message's data in a file as it's received. This file can play the same topics as they are recorded from or it can remap the existing topics too.

The main application of rosbag is data logging. The robot data can be logged and can visualize and process offline.

The rosbag command is used to work with rosbag files. Here are the commands to record and playback a bag file:

Further details about this command can be found at: http://wiki.ros.org/rosbag/Commandline

There is a GUI tool to handle the record and playback of bag files called rqt_bag. To learn more about rqt_bag, go to: http://wiki.ros.org/rqt_bag.

The ROS Master is much like a DNS server, associating unique names and IDs to ROS elements active in our system. When any node starts in the ROS system, it will start looking for the ROS Master and register the name of the node in it. So, the ROS Master has the details of all the nodes currently running on the ROS system. When any details of the nodes change, it will generate a callback and update with the latest details. These node details are useful for connecting with each node.

When a node starts publishing a topic, the node will give the details of the topic, such as name and data type, to the ROS Master. The ROS Master will check whether any other nodes are subscribed to the same topic. If any nodes are subscribed to the same topic, the ROS Master will share the node details of the publisher to the subscriber node. After getting the node details, these two nodes will interconnect using the TCPROS protocol, which is based on TCP/IP sockets. After connecting to the two nodes, the ROS Master has no role in controlling them. We might be able to stop either the publisher node or the subscriber node according to our requirement. If we stop any nodes, it will check with the ROS Master once again. This same method is used for the ROS services.

The nodes are written using the ROS client libraries, such as roscpp and rospy. These clients interact with the ROS Master using XML Remote Procedure Call (XMLRPC)-based APIs, which act as the backend of the ROS system APIs.

The ROS_MASTER_URI environment variable contains the IP and port of the ROS Master. Using this variable, ROS nodes can locate the ROS Master. If this variable is wrong, the communication between nodes will not take place. When we use ROS in a single system, we can use the IP of a localhost or the name localhost itself. But in a distributed network, in which computation is on different physical computers, we should define ROS_MASTER_URI properly; only then will the remote nodes be able find each other and communicate with each other. We need only one Master in a distributed system, and it should run on a computer in which all other computers can ping it properly to ensure that remote ROS nodes can access the Master.

The following diagram shows an illustration of how the ROS Master interacts with a publishing and subscribing node, with the publisher node publishing a string type topic with a Hello World message and the subscriber node subscribing to this topic:

Figure 8: Communication between the ROS Master and Hello World publisher and subscriber

When the publisher node starts publishing the Hello World message in a particular topic, the ROS Master gets the details of the topic and details of the node. It will search whether any node is subscribing to the same topic. If there are no nodes subscribing to the same topic at that time, both nodes remain unconnected. If the publisher and subscriber nodes run at the same time, the ROS Master exchanges the details of the publisher to the subscriber and they will connect and can exchange data through ROS messages.

When programming a robot, we might have to define robot parameters, such as robot controller gains P, I, and D. When the number of parameters increases, we might need to store them as files. In some situations, these parameters have to share between two or more programs too. In this case, ROS provides a parameter server, which is a shared server in which all ROS nodes can access parameters from this server. A node can read, write, modify, and delete parameter values from the parameter server.

We can store these parameters in a file and load them into the server. The server can store a wide variety of data types and can even store dictionaries. The programmer can also set the scope of the parameter, that is, whether it can be accessed by only this node or all the nodes.

The parameter server supports the following XMLRPC datatypes:

We can also store dictionaries on the parameter server. If the number of parameters is high, we can use a YAML file to save them. Here is an example of the YAML file parameter definitions:

/camera/name : 'nikon'  #string type 
/camera/fps : 30     #integer   
/camera/exposure : 1.2  #float  
/camera/active : true  #boolean 

The rosparam tool is used to get and set the ROS parameter from the command line. The following are the commands to work with ROS parameters:

The parameters can be changed dynamically during the execution of the node that uses these parameters, using the dyamic_reconfigure package (http://wiki.ros.org/dynamic_reconfigure).

Before running any ROS nodes, we should start the ROS Master and the ROS parameter server. We can start the ROS Master and the ROS parameter server by using a single command called roscore, which will start the following programs:

The rosout node will collect log messages from other ROS nodes and store them in a log file, and will also re-broadcast the collected log message to another topic. The /rosout topic is published by ROS nodes by using ROS client libraries such as roscpp and rospy, and this topic is subscribed by the rosout node which rebroadcasts the message in another topic called /rosout_agg. This topic has an aggregate stream of log messages. The roscore command is a prerequisite before running any ROS node. The following screenshot shows the messages printing when we run the roscore command in a Terminal.

The following is a command to run roscore on a Linux Terminal:

$ roscore

Figure 9: Terminal messages while running the roscore command

The following are explanations of each section when executing roscore on the Terminal:

The following is the content of roscore.xml:

<launch> 
 <group ns="/"> 
  <param name="rosversion" command="rosversion roslaunch" /> 
  <param name="rosdistro" command="rosversion -d" /> 
  <node pkg="rosout" type="rosout" name="rosout" respawn="true"/> 
 </group> 
</launch> 

When the roscore command is executed, initially, the command checks the command-line argument for a new port number for the rosmaster. If it gets the port number, it will start listening to the new port number; otherwise, it will use the default port. This port number and the roscore.xml launch file will pass to the roslaunch system. The roslaunch system is implemented in a Python module; it will parse the port number and launch the roscore.xml file.

In the roscore.xml file, we can see the ROS parameters and nodes are encapsulated in a group XML tag with a / namespace. The group XML tag indicates that all the nodes inside this tag have the same settings.

The two parameters called rosversion and rosdistro store the output of the rosversionroslaunch and rosversion-d commands using the command tag, which is a part of the ROS param tag. The command tag will execute the command mentioned on it and store the output of the command in these two parameters.

The rosmaster and parameter server are executed inside roslaunch modules by using the ROS_MASTER_URI address. This is happening inside the roslaunch Python module. The ROS_MASTER_URI is a combination of the IP address and port in which rosmaster is going to listen. The port number can be changed according to the given port number in the roscore command.

$ rostopic list

/rosout/rosout_agg

$ rosparam list

/rosdistro/roslaunch/uris/host_robot_virtualbox__51189/rosversion/run_id

$ rosservice list

/rosout/get_loggers/rosout/set_logger_level