DESIGN AND IMPLEMENTATION OF A SMART HOME AUTOMATION SYSTEM USING MECHATRONICS

Abstract

This bachelor's thesis presents the design and implementation of a low-cost, mechatronics-based smart home automation system tailored for a double room apartment in Mongolia. Motivated by advances in IoT, embedded systems, and the need For energy-efficient living environments, the project employs an ESP32-WROOM microcontroller as the central controller, interfaced with environmental sensors (DHT22, MQ-7, LDR, KY-037 sound sensor, ACS712 current sensor) and actuators (4-channel relay module, AC dimmer, L298N motor driver for curtains and fans, and a buzzer). The system’s firmware—developed in the Arduino IDE—implements real-time sensor polling, decision-making logic, and cloud connectivity via Sinric Pro for voice-assistant control and Node-RED/Blynk for mobile/web dashboards. Due to hardware constraints, a KY-037 sound sensor replaced a PIR module for light control, and a timed automatic shutdown was introduced to mitigate relay sticking under high loads. The prototype was systematically tested in a dormitory kitchen, achieving a 45 % reduction in idle energy consumption compared to a non-automated baseline. Voice commands like “Turn on kitchen light,” and remote toggling via mobile apps demonstrated reliable responsiveness. A comparative analysis with commercial solutions (e.g., Control4-based and Siemens-backed platforms by Moncable LLC and Digital Power LLC shows that the custom system delivers core automation features at under 10 % of the cost, highlighting its suitability for resource-constrained settings. Key contributions include a modular hardware and firmware architecture, integration of phase-angle dimming, real-time energy metering, and multi-modal control interfaces. The work confirms the hypothesis that an ESP32-driven automation system can provide energy-efficient, user-friendly smart home functionality in developingcountry contexts. Recommendations for future research encompass improved sensing (PIR or camera-based occupancy detection), solid-state switching, mesh networking, and AI-based predictive control.