Map Camera

[denysdovhan]

Map Camera

[denysdovhan]

It's well described here: https://macbury.github.io/SmartHouse/HomeAssistant/Vacuum/

[denysdovhan]

You have to specify a camera entity that streams live map of your vacuum. I guess you don't have any options because you don't have camera entities.

[CarpeDiemRo]

I have the same question. How do I have to define this entity? As far as I understood is based on generic_camera. Can you give me more detail, please?

[CarpeDiemRo]

Thanks a lot. I have doubts to root my vacuum. I will think about it.

[stibe881]

Ok. But which still_image_url i need? I tried with the follow:

http://<IP from Vacuum>/#map.html
http://<IP from Vacuum>/api/map/latest
http://<IP from Vacuum>/jpg/image.jpg

All of this doesn't work

Here ist the complete sensor. I tried with both:

camera:
  - platform: generic
    name: Saugroboter1
    still_image_url: http://<IP from Vacuum>/#map.html

  - platform: generic
    name: Saugroboter2
    still_image_url: http://<IP from Vacuum>/api/map/latest

[stibe881]

Thanks for reopening. You have any solution for my problem?

[denysdovhan]

@stibe881 I really don’t know how to help you. I decided not to flash my vacuum, since there’s no information if my model is supported. I don’t use map feature and just reimplemented that from cards I was inspired from. I haven’t had a chance to test them on a real device :(

Probably the best way to solve your problem is to ask someone in HA community forum. Or, hopefully, someone will help you in this thread.

Really sorry, but I don’t know how to help you 🙏

[tinuva]

ok I got the map funciton to work.

Here is what I did for my method:

1. Vacuum flashed with Valetudo-RE found at https://github.com/rand256/valetudo
2. Enable mqtt in /mnt/data/valetudo/config.json
3. Set up valetudo-mapper on a raspberri-pi or whatever. I have it in docker on my HA pc. https://github.com/rand256/valetudo-mapper

This will create a discovered camera entity for you in HA.

[stibe881]

ok I got the map funciton to work.

Here is what I did for my method:

1. Vacuum flashed with Valetudo-RE found at https://github.com/rand256/valetudo
2. Enable mqtt in /mnt/data/valetudo/config.json
3. Set up valetudo-mapper on a raspberri-pi or whatever. I have it in docker on my HA pc. https://github.com/rand256/valetudo-mapper

This will create a discovered camera entity for you in HA.

Hi tinuva
Thanks for your answer.
Now i have the roborock with mqtt in my home assistant.

but how can i install this? https://github.com/rand256/valetudo-mapper

can i this install on my home assistant server (ubuntu)?

i tried this. but there comes the folloe error messages when i write "npm start":

npm ERR! Linux 4.15.0-99-generic
npm ERR! argv "/usr/bin/node" "/usr/bin/npm" "start"
npm ERR! node v8.10.0
npm ERR! npm v3.5.2
npm ERR! path /root/package.json
npm ERR! code ENOENT
npm ERR! errno -2
npm ERR! syscall open

npm ERR! enoent ENOENT: no such file or directory, open '/root/package.json'
npm ERR! enoent ENOENT: no such file or directory, open '/root/package.json'
npm ERR! enoent This is most likely not a problem with npm itself
npm ERR! enoent and is related to npm not being able to find a file.
npm ERR! enoent

npm ERR! Please include the following file with any support request:
npm ERR! /root/npm-debug.log

the config file i write in my roborock vacuum on the follow path:

/mnt/data/valetudo/config.json
is this correctly?

many thanks :)

[tinuva]

ok to answer your question, yes you install it on your ubuntu server. However I would recommend running it in docker, its much easier.

Important notes, https://github.com/rand256/valetudo-mapper will only work if you used firmware from https://github.com/rand256/valetudo - the reason for this is, they split the mapper to run on a separate server instead of on the robot. Otherwise it uses up too much resources on the robot. The below I have tested on a gen1 robot using firmware vacuum_4007_valetudo_re_0_9_4.pkg

My docker compose file look like this:
docker-compose.yaml

version: '3.5'
networks:
  default:
    driver: bridge
    ipam:
      driver: default
      config:
      - subnet: 172.36.0.0/16
services:
  valetudo-mapper:
    container_name: valetudo-mapper
    restart: always
    image: rand256/valetudo-mapper:latest
    volumes:
      - /data/docker/valetudo-mapper/config.json:/app/config.json

Now to make the above work, you need to add the config in /data/docker/valetudo-mapper/config.json
(this is only for the mapper, but it should look similar to the config on your robot)

{
  "mqtt": {
    "identifier": "rockrobo",
    "topicPrefix": "valetudo",
    "autoconfPrefix": "homeassistant",
    "broker_url": "mqtt://comms.heaven.za.net",
    "caPath": "",
    "mapDataTopic": "valetudo/rockrobo/map_data",
    "minMillisecondsBetweenMapUpdates": 10000,
    "publishMapImage": true,
    "publishMapData": false
  },
  "mapSettings": {
    "colors": {
      "background": "#33a1f5",
      "background2": "#046cd4",
      "floor": "#56affc",
      "obstacle_strong": "#a1dbff",
      "path": "white",
      "forbidden_marker": "red",
      "forbidden_zone": "rgba(255, 0, 0, 0.38)",
      "cleaned_marker": "rgba(53, 125, 46, 1.0)",
      "cleaned_zone": "rgba(107, 244, 66, 0.3)",
      "cleaned_block": "rgba(107, 244, 36, 0.34)"
    },
    "drawPath": true,
    "drawCharger": true,
    "drawRobot": true,
    "drawCurrentlyCleanedZones": false,
    "drawCurrentlyCleanedBlocks": false,
    "drawForbiddenZones": true,
    "drawVirtualWalls": true,
    "scale": 4,
    "gradientBackground": true,
    "autoCrop": 20,
    "crop_x1": 30,
    "crop_y1": 70,
    "crop_x2": 440,
    "crop_y2": 440
  },
  "webserver": {
    "enabled": false,
    "port": 3000
  }
}

Config snippet file on the robot: (import part is, your identifier and topic and broker should match that of the mapper config, so that HA pick all of it automatically.

root@rockrobo:~# cat /mnt/data/valetudo/config.json
{
  "spots": [],
  "areas": [],
  "ztimers": [],
  "mqtt": {
    "enabled": true,
    "identifier": "rockrobo",
    "topicPrefix": "valetudo",
    "autoconfPrefix": "homeassistant",
    "broker_url": "mqtt://comms.heaven.za.net",
    "provideMapData": true,
    "caPath": "",
    "qos": 0
  },

Then you restart the service on the robot: service valetudo restart
Start the mapper on your server (be in same folder as your docker-compose.yaml file): docker-compose up -d && docker-compose logs -f --tail 10

Now in HA you need the following config somewhere, mine is in a package: (or do it in the web ui)

vacuum:
  - platform: mqtt

In your lovelace UI while editing this card, you should now see a new camera entity named camera.rockrobo_map if you used the above identifier + topic that I used.

[stibe881]

@tinuva
Thank you. but i dont work with docker. and i don't no how i can make it.
maybe you have time for a teamviewer session? :)

I would probably have to flash my roborock again. I suspect that valetudo is installed there and not valetudo re.

[pkishino]

you can do the same thing with valetudo. I have it working fine with valetudo.. for valetudo you need to setup https://github.com/Hypfer/ICantBelieveItsNotValetudo
then launch the npm server (you dont need the webserver, it will create an mqtt png entry) and HA will create a mqtt camera entry, and use this.

[vaproloff]

Hello everyone!

I have a problem with displaying live map using this integration.
Everything works, but Chrome tab with this lovelace card starts to load CPU to 60-80% when using map camera (from Valetudo via MQTT).

When I remove parameter 'map: camera.rockrobo_map' - problem disappears.
I have picture-entity card with the same camera entity - no problems, only within this custom card.

Update: Using Chrome dev tools I've discovered that this custom card loads camera image every 10-15 milliseconds. I suppose this process loads CPU.

Any suggestions to fix it?