Smart Lab: A Non-invasive IoT-centric Approach for Indoor Environment Automation

Abstract

This work presents a non-invasive Internet of Things (IoT)-centric framework solution for automation in indoor environments (e.g., laboratories, offices, and houses). The platform allows for Internet-connecting electronic and non-electronic devices for low-cost and fast automation control rather than replacing them as market solutions propose. Moreover, the IoT solution comprises adaptive hardware modules, a mobile application, a database, and a Message Queuing Telemetry Transport (MQTT)-based broker server deployed in a real scenario for proof of concept. The mobile application functions manage the hardware modules to remote control lamps, air conditioners, TVs, and curtains. The commands are set through predefined touchscreen buttons or voice commands interpreted by a virtual assistant based on natural language processing (NLP) and natural language understanding (NLU) machine learning algorithms embedded in a framework called Wit.ai. The database handles user subscribing by affording administration functions. Finally, a real-time broker server manages the entities' communication based on the MQTT message exchange protocol, including a security layer based on the Transport Layer Security (TLS) protocol.