The RoboCup@Home VizBox is a little webserver @Home robots can run to vizualize what is going on.

The main page shows

• an outline of the current challenge and where the robot is in the story of that challenge.
• Subtitles of what the robot and operator just said; their conversation
• Images of what the robot sees or a visualisation of the robot's world model, eg. camera images, it's map, anything to make clear what is going on to the audience.

Additionally, the server accepts HTTP POSTs to which a command sentence can be submitted on /command

The server abstracts over the underlying robot via a Backend. A backend accepts messages from the robot's internals. A message is forwarded to the web page via websockets.

Currently, only a ROS backend is implemented:

• operator_text std_msgs/String What the robot has heard the operator say
• robot_text std_msgs/String What the robot itself is saying
• challenge_step std_msgs/UInt32 Active item index in the plan or action sequence the robot is executing.
• image sensor_msgs/Image Image of what the robo sees or anything else interesting to the audience

• command std_msgs/String Command HTTP POSTed to the robot.

• Allow robot to push action sequence and challenge name to server. Allows for GPSR action sequences etc.

git clone [email protected]:WalkingMachine/vizbox.git
cd vizbox
pip install -r requirements.txt --user

roscore # in separate terminal
./server.py image:=/usb_cam/image_raw # Remaps the image-topic to output of the USB cam, see below

Open The web page on localhost

To reproduce the the screenshot:

roslaunch usb_cam usb_cam-test.launch # separate terminal
rostopic pub /challenge_step std_msgs/UInt32 "data: 0" --once
rostopic pub /robot_text std_msgs/String "data: 'Hello operator'" --once
rostopic pub /operator_text std_msgs/String "data: 'Robot, follow me'" --once
rostopic pub /robot_text std_msgs/String "data: 'OK, I will follow you'" --once;
rostopic pub /challenge_step std_msgs/UInt32 "data: 1" --once