A project to enable MQTT control for BLE adjustable beds. Currently supported adjustable beds are:
- Serta adjustable beds with Bluetooth, like the Serta Motion Perfect III
- 'Glide' with the jiecang BLUE controller (Dream Motion app)
- 'A H Beard' with the DerwentOkin BLE controller ("Comfort Enhancement 2" aka "Comfort Plus" app)
- 'HankookGallery' with the DerwentOkin BLE controller OKIN Comfort Bed app - (It maybe old version of 'Compfort Plus' app)
- Linak KA20IC with the "Bed Control" app
First, you need to install bluepy from your package manager of choice. On a pi running Debian, you can use the following apt command to install Poetry, bluez and its dependencies:
sudo apt update
sudo apt install python3-pip pipx bluez libglib2.0-dev
pipx install poetry
pipx ensurepathYou will need to log out and log back in for the PATH changes to take effect. You can then install the necessary Python dependencies:
cd mqtt-bed
poetry installThis will automatically install all of the necessary dependencies outlined in the pyproject.toml file. If you prefer not to use Poetry, you can still do the typical pip3 install . but will need to maintain your own virtual environment.
To configure all of the settings used in the program, such as the MQTT credentials, bluetooth address of your bed, etc, you will need to modify the values in config.yaml. This setup is essential to getting your integration functional.
There are numerous ways to get the address for your bed. If you can already control the bed via your phone, check the list of paired bluetooth devices, and find the address for your bed's connection in the format 00:00:00:00:00:00.
If you have a Serta bed, and are using the Serta MP Remote app, you can find the mac address there.
Otherwise, you can use hcitool from the bluez package to find the address of your bed:
$ hcitool lescan
LE Scan ...
61:61:61:4E:A0:B0 (unknown)
5F:E4:19:C4:13:CC (unknown)
6F:40:F7:4A:8E:23 (unknown)
7C:EC:79:FF:6D:02 (unknown)
7C:EC:79:FF:6D:02 base-i4.00000233Once you have the address for your bed, you will need to fill that into the BED_ADDRESS variable in the config.yaml file.
To run the program in the poetry virtual environment, you can run:
poetry run python /path/to/mqtt-bed/mqtt-bed.pyYou will need to substitute the correct path to the program source depending on where you cloned the repository.
The program has one optional argument, which will control the verbosity of the program. By default, the log level is set to INFO but can be easilt changed with the --log option. When doing development or troubleshooting, it is recommended to runm with --log DEBUG
usage: mqtt-bed.py [-h] [--log LOG_LEVEL]
BLE adjustable bed control over MQTT
options:
-h, --help show this help message and exit
--log LOG_LEVEL Set the log level (DEBUG, INFO, WARNING, ERROR, CRITICAL)If you are using a dedicated device (like a Raspberry/Orange Pi Zero W) to run this program, you will likely want to create a Systemd service to run this application at startup, and restart after crashes. To do this on a Debian-based system, you will need to create a file called /etc/systemd/system/mqtt-bed.service, and paste the following contents:
[Unit]
Description=mqtt-bed service
After=network.target
[Service]
ExecStart=/home/pi/.local/bin/poetry run python /home/pi/mqtt-bed/mqtt-bed.py
WorkingDirectory=/home/pi/mqtt-bed
User=pi
Restart=always
RestartSec=1
[Install]
WantedBy=multi-user.targetIn this file, you may need to swap out the username in the User variable, as well as the paths used in ExecStart and WorkingDirectory. The ExecStart command should contain:
/path/to/poetry run pythonwhich can be found by runningwhich poetry- The absolute path to the
mqtt-bed.pyfile.
After creating this file, you will need to run sudo systemctl daemon-reload to grab the new service file, sudo systemctl enable mqtt-bed to run the service on boot, and sudo systemctl start mqtt-bed to start the service for the first time.
You can check the status of the service at any time by running sudo systemctl status mqtt-bed.
You can also view the live logs from this service at any time by running sudo journalctl -u mqtt-bed.service -f
The YAML used in Home Assistant to integrate your bed will vary by your installation and bed type, but you can find example YAML for in the homeassistant-script.yaml file in this repository.
In addition, it is common to run the MQTT Broker service in your Home Assitant installation if you do not have a broker running already, in which case you can use the official Mosquito broker addon. Once you have your broker setup, you will want to fill in your Home Assistant information in the MQTT config section of the config.yaml file.
For an even more seamless integration with Home Assistant, this project supports MQTT Discovery, which allows certain bed controllers to be automatically recognized and configured in Home Assistant. This feature simplifies the setup process and reduces the need for manual YAML configuration.
To enable MQTT Discovery, set the MQTT_DISCOVERY option to true in the config.yaml file.
Note: MQTT Discovery is currently only supported by the Linak controller.
If you are running this application on a dedicated device, you likely need to pair and trust the bed on your device.
First, you need to put your bed into pairing mode. The method to do this will be highly dependent on your bed type.
Once it is in pairing mode, you can run the following bluetoothctl commands to enable BLE, scan for the device, pair, and trust the bed:
bluetoothctl
[bluetooth]# power on
[bluetooth]# agent on
[bluetooth]# default-agent
[bluetooth]# scan onNow, you should be able to find your bed in the list of discovered devices. For example:
[NEW] Device 00:00:00:00:00:00 BedNow you'll want to initiate pairing with the device, connect once, trust the device, and disconnect. You will need to substitute in the correct address for your device:
[bluetooth]# pair 00:00:00:00:00:00
[bluetooth]# connect 00:00:00:00:00:00
[bluetooth]# trust 00:00:00:00:00:00
[bluetooth]# disconnect 00:00:00:00:00:00
[bluetooth]# exitIf your adjustable bed is not integrated into this repository yet, and you create your integration, it would be great to add your controller to this project for other to use!
To integrate your own bed, you should follow the examples in controllers/dewertokin.py and controllers/linak.py utilizing the bluepy package rather than the deprecated pygatt/gatttool integrations.
Just create your own controller class with an __init__ function to kick off the connection to your bed, and add your controller to the list of valid controllers in the main() function of mqtt-bed.py.