Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

HKD-SHO: A hybrid smart home system based on knowledge-based and data-driven services

Mingming Qiu, Elie Najm, Rémi Sharrock, Bruno Traverson

TL;DR

The paper addresses the challenge of building smart home services that are both explicable and adaptive. It proposes HKD-SHO, a hybrid system that fuses knowledge-based rules with data-driven reinforcement learning, enriched by a PBRE variant that extracts and generalizes rules from learning outcomes. The architecture combines a knowledge representation layer, rule management, state proposition mechanisms, and a decision maker, enabling dynamic yet explainable actuator control via EPbA/RSAbA MARL implementations. Comparative simulations show HKD-SHO generally outperforms purely learning-based and purely rule-based baselines, demonstrating improved robustness and user satisfaction while maintaining explainability. The work highlights practical potential for deploying hybrid, dynamically adapting smart-home services and points to future real-world validation and efficiency improvements.

Abstract

A smart home is realized by setting up various services. Several methods have been proposed to create smart home services, which can be divided into knowledge-based and data-driven approaches. However, knowledge-based approaches usually require manual input from the inhabitant, which can be complicated if the physical phenomena of the concerned environment states are complex, and the inhabitant does not know how to adjust related actuators to achieve the target values of the states monitored by services. Moreover, machine learning-based data-driven approaches that we are interested in are like black boxes and cannot show the inhabitant in which situations certain services proposed certain actuators' states. To solve these problems, we propose a hybrid system called HKD-SHO (Hybrid Knowledge-based and Data-driven services based Smart HOme system), where knowledge-based and machine learning-based data-driven services are profitably integrated. The principal advantage is that it inherits the explicability of knowledge-based services and the dynamism of data-driven services. We compare HKD-SHO with several systems for creating dynamic smart home services, and the results show the better performance of HKD-SHO.

HKD-SHO: A hybrid smart home system based on knowledge-based and data-driven services

TL;DR

The paper addresses the challenge of building smart home services that are both explicable and adaptive. It proposes HKD-SHO, a hybrid system that fuses knowledge-based rules with data-driven reinforcement learning, enriched by a PBRE variant that extracts and generalizes rules from learning outcomes. The architecture combines a knowledge representation layer, rule management, state proposition mechanisms, and a decision maker, enabling dynamic yet explainable actuator control via EPbA/RSAbA MARL implementations. Comparative simulations show HKD-SHO generally outperforms purely learning-based and purely rule-based baselines, demonstrating improved robustness and user satisfaction while maintaining explainability. The work highlights practical potential for deploying hybrid, dynamically adapting smart-home services and points to future real-world validation and efficiency improvements.

Abstract

A smart home is realized by setting up various services. Several methods have been proposed to create smart home services, which can be divided into knowledge-based and data-driven approaches. However, knowledge-based approaches usually require manual input from the inhabitant, which can be complicated if the physical phenomena of the concerned environment states are complex, and the inhabitant does not know how to adjust related actuators to achieve the target values of the states monitored by services. Moreover, machine learning-based data-driven approaches that we are interested in are like black boxes and cannot show the inhabitant in which situations certain services proposed certain actuators' states. To solve these problems, we propose a hybrid system called HKD-SHO (Hybrid Knowledge-based and Data-driven services based Smart HOme system), where knowledge-based and machine learning-based data-driven services are profitably integrated. The principal advantage is that it inherits the explicability of knowledge-based services and the dynamism of data-driven services. We compare HKD-SHO with several systems for creating dynamic smart home services, and the results show the better performance of HKD-SHO.
Paper Structure (24 sections, 6 equations, 13 figures)

This paper contains 24 sections, 6 equations, 13 figures.

Table of Contents

  1. Introduction
  2. Related work
  3. Proposed HKD-SHO system
  4. Rule management
  5. Rule extraction
  6. Rule deletion
  7. State proposition
  8. Machine learning-based system
  9. EPbA
  10. RSAbA
  11. Extracted rules reasoner
  12. Decision maker
  13. Working process of HKD-SHO
  14. Comparative experiment
  15. Simulated environment
  16. ...and 9 more sections

Figures (13)

  • Figure 1: HKD-SHO architecture
  • Figure 2: principle of PBRE variant
  • Figure 3: Architecture of EPbA
  • Figure 4: Architecture of RSAbA
  • Figure 5: Detailed working process of HKD-SHO
  • ...and 8 more figures