Realizing Space-oriented Control in Smart Buildings via Word Embeddings

TL;DR

This paper introduces a space-oriented control framework for smart buildings that uses natural language processing and word embeddings to enable intuitive, room-level control via a chat-based GUI, improving development efficiency and portability.

Contribution

The paper presents a novel space-oriented control framework utilizing NLP and word embeddings, with a chat-based GUI and vector database integration for smart building management.

Findings

Prototype tested at University of Tokyo successfully demonstrates feasibility.

Effective conversion of natural language to building API calls achieved.

Improved control intuitiveness and efficiency over traditional device-oriented methods.

Abstract

This paper presents a novel framework for implementing space-oriented control systems in smart buildings. In contrast to conventional device-oriented approaches, which often suffer from issues related to development efficiency and portability, our framework adopts a space-oriented paradigm that leverages natural language processing and word embedding techniques. The proposed framework features a chat-based graphical user interface (GUI) that converts natural language inputs into actionable OpenAI API calls, thereby enabling intuitive space level (e.g., room) control within smart environments. To support efficient embedding-based search and metadata retrieval, the framework integrates a vector database powered by Elasticsearch. This ensures the accurate identification and invocation of appropriate smart building APIs. A prototype implementation has been tested in a smart building environment at the University of Tokyo, demonstrating the feasibility of the approach.