This repository contains the source code for the following publication:
Georg Götz, Abraham Martinez Ornelas, Sebastian J. Schlecht and Ville Pulkki, "Autonomous Robot Twin System for Room Acoustic Measurements", Journal of the Audio Engineering Society, Vol. 69, No. 4, April 2021.
Please find the companion page here: http://research.spa.aalto.fi/publications/papers/artsram/
Please find the published paper here: https://www.aes.org/e-lib/browse.cfm?elib=21033
The Autonomous Robot Twin System for Room Acoustic Measurements (ARTSRAM) is capable of measuring room impulse responses (RIRs) with variable sound source and receiver positions. It consists of two independent robots that are able to move freely in a room. Both robots are equipped with collision sensors, thus allowing them to explore the room autonomously. The measurements of RIRs are complemented with corresponding position information of the robots.
The base for each robot is an iRobot Create 2 Roomba robot, which is controlled by a Raspberry Pi single-board computer. HTC Vive trackers of the second generation are used for tracking the position of the robots. Although it would be possible to mount several microphone arrays and loudspeakers on each of the robots to enable multi-way RIR measurements, we chose to clearly distinguish between a source and a receiver robot in this paper. The source robot uses a Minirig MRBT-2 portable loudspeaker to play back excitation signals that are recorded by the receiver robot with a Zoom H3-VR first-order microphone array. The resulting RIRs are stored on a microSD card inserted into the receiver robot's Raspberry Pi.
The entire measurement procedure is implemented in Python. It is controlled by a main measurement script running on a separate measurement laptop. The main measurement script sends commands over a TCP network socket to the Raspberry Pis of the source and receiver robot. Subsequently, the corresponding server scripts running on the Raspberry Pis handle the commands. For example, the server scripts can trigger robot movements or an RIR measurement. The HTC Vive trackers directly communicate with the measurement laptop over another wireless connection, thus allowing the main measurement script to immediately access and store the positions of both robots.
Usually, the following procedure has to be followed to start the measurement procedure:
- Adapt all parameters in measurement_params.py to your setup, especially IP addresses around device names are crucial for the script to work properly. Place robots in middle of the room, directly facing each other and as close as possible
- Turn on the Roomba iCreate 2 robots, microphone array and loudspeaker. Connect the Raspberry Pis to power.
- Execute robo_socket_client.py on your measurement laptop.
- Execute robo_socket_server.py on the Raspberry Pis and provide as an argument either 1 or 2 to tell the script which robot is receiver or source respectively.
- Turn on one Vive Tracker after another.
Most of these steps will also appear as prompts on your measurement laptop after executing robo_socket_client.py.
- HTC Vive base stations outside play area
- Facing approx 45 downwards
- Room Calibration
- Make sure that HTC Vive headset is inside the play area
- Use advanced mode to put points manually and have straight edges
- Go around the base stations with the controllers and look if the controller position coincides with the base station position on the computer screen if you have them at the same location, otherwise something is wrong with the base station placement or the previous steps
- Ensure that the form you marked is somewhat close to your play area, otherwise do the previous steps again or move base stations a little.
- Make a screen shot of the calibration window with the corner points of the play area
- Make sure that all devices are placed correctly on the robot
- Microphone should have the Zoom logo at the front of the robot (i.e. towards forward driving direction)
- Trackers should have status led at back of the robot (i.e.towards backward driving direction)
- Power banks should be attached at the front of the robot, because they are good for stability
- Connect all devices with the Raspberry Pis
- Turn on Microphone array and put it into I/O Mode with Ambisonics (e.g. B-Format, FuMa) settings, write down which convention you picked
- Make sure all IP addresses in the scripts are correct
- Run ifconfig -a on the Pis or ipconfig -all on a Mac measurement laptop to find out
- Set the max straight drive times according to room dimensions
- Adapt sweep length according to room
- Set max RIR length according to room
- Set the amount of desired measurements
- Turn on the HTC Vive connector box
- Measure room corners with the trackers mounted on the robots, this is crucial for determining the coordinate system later
- Measure positions that would span your x and y axis
- Measure orientation angles of robots
- Place robots in middle of the room, directly facing each other and as close as possible
- Turn on robots by pressing the button
- Light should be green, usually it says “English”
- Prepare MicroSD card with Raspberry Pi Imager. Just use standard RapsberryPi OS
- Plug all cables (usb keyboard and mouse, HDMI) to Pi and then the power cable
- Carefully go through setup
- It will take 2-3 times until all updates are done
- For me, there were both times errors, requiring me to additionally run sudo dpkg --configure -a
- Go to RaspberryPi Configuration and change the hostname from raspberrypi to pi
- To enable connecting via ssh and Remote Desktop:
- Enable SSH and VNC in configuration sudo apt install tightvncserver -y sudo apt --fix-broken install sudo apt install xrdp -y
- Set up Code and measurement folder from git
- python3 -m pip install -r requirements.txt
- sudo apt-get install python-dev libatlas-base-dev
- sudo apt-get install libasound-dev
- sudo apt-get install libportaudio2