Raspberry Pi jako inteligentní centrum chytré domácnosti pro osoby se specifickými potřebami

Abstract

This bachelor's thesis presents the design and implementation of a smart home system intended for individuals with limited motor function or reduced communication abilities. The central unit is a Raspberry Pi 5 running an Apache web server with a PHP~8 backend and a MariaDB database. The system allows users to capture and replay infrared signals in two modes -- protocol-based (NEC, RC-5, Sony SIRC) and as raw pulse timing sequences -- as well as 433~MHz RF signals using direct raw waveform replay. It also supports sending HTTP commands to IoT devices on the local network and dispatching SMS notifications via an SMS gateway. For users with severe motor limitations, two alternative input interfaces are implemented: an HX710B pressure sensor responding to three breath gestures, and two external buttons connected via 3.5~mm jack connectors. A galvanically isolated relay output is included for nurse call functionality. The entire solution operates autonomously on the local network without dependency on cloud services. The outcome is a functional prototype featuring a ten-table database model, a REST-like PHP API, and Python daemons managed via systemd, all controlled through a responsive web interface built on Bootstrap~5.