Skip to content

Configuration

Area Occupancy Detection is configured entirely through the Home Assistant user interface.

Adding a New Area

  1. Navigate to Integrations: Go to Configuration -> Devices & Services -> Integrations.
  2. Add Integration: Click the + Add Integration button in the bottom right.
  3. Search: Search for "Area Occupancy Detection" and select it.
  4. Configure Area Name:
    • Enter a descriptive Name for the area this instance will monitor (e.g., "Living Room", "Office"). This name will be used in entity IDs.

Configuration Options

Area Purpose

The first step after naming the area is choosing its purpose. This sets a sensible default for the decay half life used when probability decreases. The purpose selection affects how quickly the system "forgets" about occupancy after activity stops.

Purpose Description Default Half-Life
Passageway Quick walk-through: halls, stair landings, entry vestibules. Motion evidence should disappear almost immediately after the last footstep. 60 seconds
Utility Laundry room, pantry, boot room. Short functional visits (grab the detergent, put on shoes) with little lingering. 120 seconds
Food-Prep Kitchen work zone around the hob or countertop. Residents step away to the fridge or sink and return; a few minutes of memory prevents flicker. 300 seconds
Eating Dining table, breakfast bar. Family members usually stay seated 10-20 minutes but may be fairly still between bites. 600 seconds
Working / Studying Home office, homework desk. Long seated sessions with occasional trips for coffee or printer; ten-minute half-life avoids premature "vacant". 600 seconds
Social Living room, play zone, game area. Conversations or board games create sporadic motion; evidence fades gently to ride out quiet pauses. 720 seconds
Relaxing TV lounge, reading nook, music corner. People can remain very still while watching or reading; a quarter-hour memory keeps the room "occupied" through stretches of calm. 900 seconds
Sleeping Bedrooms, nap pods. Motion is scarce; a long half-life prevents false vacancy during deep sleep yet lets the house revert to "empty" within a couple of hours after everyone gets up. 1800 seconds

You can override the resulting half life in the parameters section if needed.

After providing the name and purpose, you'll be guided through selecting sensors and configuring parameters. You can also reconfigure these later by clicking Configure on the integration card.

Sensor Selection

You will be prompted to select entities for various categories. You only need to select sensors relevant to the specific area you are configuring.

Sensor Type Entity Type Description Default States/Range
Primary Occupancy Sensor (Required) binary_sensor One reliable motion or occupancy sensor. Crucial as ground truth for Prior Probability Learning. on
Motion Sensors binary_sensor Additional motion sensors in the area such as PIR or mmWave sensors. on
Door Sensors binary_sensor Relevant door sensors. Closed
Window Sensors binary_sensor Relevant window sensors. Open
Media Devices media_player Relevant media players. playing, paused
Appliances switch, binary_sensor, sensor Relevant switch or sensor entities representing appliances. on, standby
Illuminance Sensors sensor Illuminance sensors measuring light levels (lux) 30.0 - 100000.0
Temperature Sensors sensor Temperature sensors measuring temperature 18.0 - 24.0
Humidity Sensors sensor Humidity sensors measuring humidity 70.0 - 100.0

Available States

Different sensor types support different state configurations:

Door Sensors

  • Open - Door is open
  • Closed - Door is closed (default)

Window Sensors

  • Open - Window is open (default)
  • Closed - Window is closed

Media Devices

  • Playing - Media is actively playing (default)
  • Paused - Media is paused
  • Idle - Media device is idle
  • Off - Media device is off

Appliances

  • On - Appliance is active (default)
  • Off - Appliance is off
  • Standby - Appliance is in standby mode

Motion Sensors

  • Active - Motion detected (default)
  • Inactive - No motion detected

Parameters

Parameter Description Range Default
Occupancy Threshold (%) The probability percentage required for the main Occupancy Status binary sensor to turn on 1-99 50
History Period (Days) The number of past days to analyze when performing Prior Probability Learning 1-90 7
Decay Enabled Toggle whether to enable the Probability Decay feature True/False Enabled
Decay Half Life (Seconds) When decay is enabled this defines how long it takes for the occupancy probability to reduce by half after activity stops 10-3600 Based on purpose
Motion Timeout (Seconds) How long motion sensors remain "active" after detecting motion before automatically resetting to inactive 0-3600 300 (5 minutes)

Wasp in Box Feature

The "Wasp in Box" is a virtual sensor that combines motion and door sensor data to create a more reliable occupancy indicator. It's particularly useful for areas where people might be still for extended periods.

Parameter Description Range Default
Wasp in Box Enabled Toggle the Wasp in Box virtual sensor True/False Disabled
Motion Timeout How long motion sensors remain active after detecting motion 0-3600 seconds 300 seconds
Weight Influence of the Wasp in Box sensor on final probability calculation 0.0-1.0 0.8
Max Duration Maximum time the Wasp in Box can indicate occupancy without new motion 0-86400 seconds 3600 seconds (1 hour)

Sensor Weights

Adjust the influence of different types of sensors on the final probability calculation. Weights range from 0.0 (no influence) to 1.0 (maximum influence). Default values are provided based on typical sensor reliability for occupancy.

Sensor Type Default Weight
Motion Sensor 1.00
Media Device 0.70
Appliance 0.40
Door Sensor 0.30
Window Sensor 0.20
Environmental Sensor 0.10
Wasp in Box 0.80

Reconfiguring an Existing Area

  1. Go to Configuration -> Devices & Services -> Integrations.
  2. Find the Area Occupancy Detection integration card for the area you want to change.
  3. Click Configure.
  4. You can then step through and modify any of the sensor selections, parameters, or weights.

Click Submit on each step to save changes.

Automatic Learning

The integration automatically:

  • Analyzes historical data to determine sensor reliability
  • Calculates correlation between sensor states and occupancy
  • Adjusts probabilities based on learned patterns
  • Updates calculations in real-time

Created Entities

After configuration, the integration creates:

  1. Occupancy Probability Sensor – Shows the calculated probability as a percentage.
  2. Occupancy Status Binary Sensor – Indicates if the area is occupied based on the threshold.
  3. Prior Probability Sensor – Displays the combined prior used for calculations.
  4. Evidence Sensor – Lists entities providing evidence and those that are inactive.
  5. Decay Status Sensor – Indicates progress of probability decay.
  6. Occupancy Threshold Number – Allows adjusting the threshold used by the binary sensor.
  7. Wasp in Box Sensor – Virtual sensor combining motion and door data (if enabled).

Adjusting Configuration

You can adjust settings anytime through the integration options:

  1. Go to Settings → Devices & Services
  2. Find Area Occupancy Detection
  3. Click "Configure"
  4. Modify settings as needed

Next Steps

After configuration:

  1. Monitor the created entities to ensure they reflect actual occupancy
  2. Adjust the threshold if needed
  3. Review the Basic Usage guide