KNX infrared presence sensor

New project in Wrocław

Applications Used

Lighting

Heating, Ventilation & Air conditioning (HVAC)

Security and Safety

Operation and Visualisation

Energy Management/Smart Metering

Other

Automation and Remote Access

Blind and Shutter Control

Project Details

Category Value
Project Category New project
Project Type Other
Year of Completion 2026
Project Size 1 – 100
Number of KNX Devices Used 1 – 25
KNX Transmission Media Twisted pair
Man Hours 1 – 100
Man Power 1 – 5

Featured in this project

Weinzierl Engineering GmbH

Detailed Description

Energy Efficiency

The usage of automated functions already save energy in every installation. Especially the usage of KNX has a lot of potential for further energy savings, which can contribute to a better environmental footprint. Be extending KNX to more functions and applications, energy savings of up to 60% can be realised. Thanks to the flexibility of KNX, these functions can easily be implemented in this installation

Innovation

Smart homes and buildings are always an upgrade in comparison to a conventional installation. Not only automated functions and a higher level of comfort contribute to a higher living standard, but also the possibility to be ready for coming innovations. Thanks to the usage of KNX, the door has been opened to integrate new innovative solutions, using the worldwide STANDARD for home and building control.

Applications Used

  • Lighting - Switching, Light Scenes, Automatic Occupancy Detection
  • Heating, Ventilation & Air conditioning (HVAC) - Automatic Occupancy Detection, Individual Room Control / Zone Control
  • Security and Safety
  • Operation and Visualisation
  • Energy Management/Smart Metering
  • Other
  • Automation and Remote Access
  • Blind and Shutter Control

Customers Description

This project is a working prototype of a KNX-integrated presence sensor based on thermal image analysis. The system uses a Waveshare Thermal Camera HAT connected to a Raspberry Pi, providing 80 × 62 px infrared images with a 45° field of view. The images are processed locally by a convolutional neural network trained on my own labelled dataset. At the current stage, the model classifies the detection zone into three occupancy classes: 0, 1 or 2 people. The processed information is transferred to the KNX installation via twisted pair using a Weinzierl BAOS interface. The device sends both a binary presence signal and the detected number of people. This enables more intelligent control of lighting and HVAC systems, for example adjusting heating or ventilation according to real occupancy. Since the system uses thermal data and performs local analysis without storing images, it provides a higher level of privacy than standard camera-based solutions.

Contact Information

Company Name Automatyka Domu Sp. z o.o.
Company Email nikodem.leleniewski@automatykadomu.com
Company Website https://automatykadomu.com
Contact Name Nikodem Leleniewski
Contact Function System Integrator