Ovládání chytré domácnosti pomocí gest detekovanými radary

Abstract

This bachelor thesis deals with the development and implementation of a system for controlling a smart home using radar-detected gestures, responding to the need to simplify the interaction between humans and technology. In the theoretical part, the thesis analyses the issue of smart homes and radar technologies for contactless gesture detection. In the development of the system, a combination of data processing and evaluation in TLV 1051 format and integration with KNX system for communication with smart home was used. The IWR6843 radar module from Texas Instruments was applied to control home devices such as lighting and air conditioning. All the software components were implemented in Python, and an application with a graphical user interface was developed in the Tkinter and customTkinter libraries, which is also executable via an exe file, making it easy to use. Quantitative methods were applied to determine a probabilistic threshold for successful gesture detection, and the most reliable gestures for control were recommended. This work not only provides significant improvements in quality of life for people with health limitations, but is also suitable for environments where minimal physical contact is preferred, such as medical facilities or laboratories. Although successes have been made in gesture recognition, there is still room for improvement. For future research, it is recommended to extend the system to train its own more robust neural network to increase the accuracy and reliability of gestures.