PlantBot

Smart tracking for when to water plants with a pesky SlackBot!

• Prerequisites
• Installation
• Configuration
• Deployment
• Slack Output

Prerequisites

PlantBot requires at minimum the following hardware setup:

• Raspberry Pi Zero W for remote compute and storage needs
• Xiaomi Mi Flora sensor(s) to monitor temperature, soil moisture, conductivity and light conditions

Installation

Installation of PlantBot requires a working Raspberry Pi. This guide is suggested for setting up a Pi in headless mode.

PlantBot can be installed by cloning this repo and its python requirements.

$ git clone https://github.com/jfri3d/PlantBot.git
$ cd PlantBot
$ pip3 install -r requirements.txt

Note that all processing is time-based (i.e. with cron), requiring the correct local time on the pi.

sudo cp /usr/share/zoneinfo/<continent>/<city> /etc/localtime

Configuration

Minimal configuration with a .envrc file based on .envrc.template is required in order to link with Slack and determine correct sunrise/sunset times (i.e. based on location).

• build a slackbot
• define a giphy "app"
• identify the accurate geolocation

export SLACK_TOKEN=
export SLACK_BOT_TOKEN=
export GIPHY_KEY=
export LAT=
export LON=

Finally, the most important aspect is to link the hardware with the software. In other words, determine the MAC address of each respective Mi Flora sensor. The basic linux tool hcitool identifies everything on the current network. The Flower Care sensor identifies the MAC address(es) that need to be saved.

$ sudo hcitool lescan

LE Scan ...

AA:AA:AA:AA:AA:AA Flower care
UU:UU:UU:UU:UU:UU (unknown)
XX:XX:XX:XX:XX:XX Flower care

The saved MAC address(es) need to be saved into the ./data/plant_def.json file containing the following schema. Note that the default definition file naming can be changed in ./scripts/constants.py.

{
  "plants": [
    {
      "name": "minty",
      "mac_address": "XX:XX:XX:XX:XX:XX",
      "min_moisture": 30
    },
    {
      "name": "oregano",
      "mac_address": "AA:AA:AA:AA:AA:AA",
      "min_moisture": 20
    }
  ]
}

Deployment

The two running processes for running PlantBot are as follows:

• ./scripts/plantbot.py - responsible for daily scheduling of "reading" raw plant measurements
• ./scripts/slackbot_alert.py - responsible for alerting via Slack which plant needs to be watered

These processes are deployed with supervisor, see this for installation and setup. The end result should be an adapted /etc/supervisor/supervisord.conf file as such:

[unix_http_server]
file=/var/run/supervisor.sock   ; (the path to the socket file)
chmod=0700                       ; sockef file mode (default 0700)

[supervisord]
logfile=/var/log/supervisor/supervisord.log ; (main log file;default $CWD/supervisord.log)
pidfile=/var/run/supervisord.pid ; (supervisord pidfile;default supervisord.pid)
childlogdir=/var/log/supervisor            ; ('AUTO' child log dir, default $TEMP)

[rpcinterface:supervisor]
supervisor.rpcinterface_factory = supervisor.rpcinterface:make_main_rpcinterface

[supervisorctl]
serverurl=unix:///var/run/supervisor.sock ; use a unix:// URL  for a unix socket

[include]
files = /home/pi/PlantBot/supervisor/*.conf

• Controlling supervisor processes (first read from newly adapted *.conf)

sudo supervisorctl reread
sudo supervisorctl

• Adding new processes (located in ./supervisor/)

supervisor> add plantbot alert

• Check status of processes

supervisor> status

plantbot      RUNNING   pid 1233, uptime 0:00:04
alert         RUNNING   pid 1234, uptime 0:00:04

Slack Output