NMBURobotics/ros2_python_demos

Purpose

This repo provides example python nodes running on ROS2. The developments were done under an Ubuntu 20.04 machine running ROS2 Foxy version.

The demo packages showcases many of essential concept in ROS-ROS2 such as;

• Topic Publisher/Subscriber
• Service Server/Clients
• Action Server/Clients
• Parameter loading and Launch files
• Custom message definition and usage (see ros2_demo_custom_msgs)

Build

Assuming that ROS2 Foxy desktop is installed on your machine, create a workspace and build the workspace with colcon;

mkdir -p ros2_ws/src
cd ros2_ws
colcon build --symlink-install

You can test a sample node (e.g. the topic publisher and subscriber)

cd ros2_ws
source /opt/ros/foxy/setup.bash
source install/setup.bash
ros2 launch ros2_demo_python_nodes start_topic_pub_sub.launch.py

Additional depenedencies

Some nodes might require installation of 3rd party libraries such as gTTS (for generating speech from text). You should be able to install other dependencies them with pip3 in case you get "no module XX found" errors.

Action and Service Nodes

Under ros2_demo_python_nodes/ros2_demo_python_nodes/text_to_speech_service_server.py, We provide two service servers to generate speech.

For instance;

In a terminal;

cd ros2_ws
source /opt/ros/foxy/setup.bash
source install/setup.bash
ros2 launch ros2_demo_python_nodes start_speech_server.launch.py

In a seperate terminal

source /opt/ros/foxy/setup.bash
source install/setup.bash
ros2 service call /speak_random std_srvs/srv/Trigger

A randomly picked sentence (Loaded from params.yaml)will be played back.

Or you can generate the speech for a specific sentence;

source /opt/ros/foxy/setup.bash
source install/setup.bash
ros2 service call /speak_phrase ros2_demo_custom_msgs/srv/Phrase "{phrase: 'I will be Spoken'}"

The action nodes might come up as scary if you are a beginner, It is suggested you get grasp of ROS Actions in order to test and understand navigate_to_pose action interface provided in ros2_demo_python_nodes/ros2_demo_python_nodes/text_to_speech_service_server.py.

THe purpose of this action node is to drive a robot to a certain pose (hence navigate_to_pose), It is required that you have a valid tf tree, depending on your localization setup you should have either in map or odom frame to base_link transforms, modify ros2_demo_python_nodes/ros2_demo_python_nodes/helpers.py with correct frame id (either map or odom) in the get_curr_robot_pose function.

To test the actions, source your workspace and do;

ros2 launch ros2_demo_python_nodes start_nav2pose_action.launch.py

In a seperate terminal;

ros2 run ros2_demo_python_nodes nav_to_pose_action_client

For this to succeed, you do need a complete tf transfrom from odom/map to base_link so that robot can know its current position and navigate itself towards the goal. You can change the goal pose in ros2_demo_python_nodes/ros2_demo_python_nodes/nav_to_pose_action_client.py to send robot to differnt pose.