Abstract

Intelligent home automation systems have gained prominence with advancements in IoT technologies, yet many lack adaptability to individual user preferences. This thesis presents HomeOrbit, a modular smart home prototype designed to address this gap through personalized automation. The system integrates an ESP32 microcontroller with environmental sensors (DHT11 for temperature/humidity, a photoresistor for ambient light) and actuators (LEDs) to enable context-aware responses. A multi-layered software architecture employs MQTT for efficient data communication and a web-based GUI for real-time monitoring and manual control. While the current implementation utilizes rule-based logic (e.g., activating "Coffee Mode" under specific temperature/humidity thresholds), the framework explores future integration of machine learning for adaptive behavior. Key contributions include a scalable hardware design, seamless sensor-actuator coordination, and a user-centric interface. The prototype demonstrates the feasibility of personalized automation, and lays groundwork for expanding IoT interoperability and offers recommendations and ideas on further scalibility and extensibility.

Advisor

Visa, Sofia

Department

Computer Science

Publication Date

2025

Degree Granted

Bachelor of Arts

Document Type

Senior Independent Study Thesis

Download

Share

COinS
 

© Copyright 2025 RM Shahriar Hoque