Digital Twin Models Supporting Cognitive Buildings for Ambient Assisted Living

Author:

Corneli Alessandra,Binni Leonardo,Naticchia Berardo,Vaccarini Massimo

Abstract

AbstractThe rapid and global aging of population is outlining the need for environments that can provide support for these individuals during their daily activities. The challenge of an aging society is being addressed through the incorporation of new technologies into the home environment, which is nothing less than Ambient Assisted Living (AAL). To date, some of the AAL solutions exploit AI models to recognize the elderly’s behaviors through data collected by sensors. In recent times, Digital Twins (DTs) at building level have begun to appear on the construction domain. These are still under development but through the integration of users into assessments, they improve efficiency, prevention, and prediction of likely events through real-time AI computing. The integration of DT and AAL defines cognitive buildings which aim to learn at scale, reason with a purpose, and co-operate with users in a natural way. This research aims to develop DT models to achieve scenario awareness to provide support to elderly people living alone and suffering from cognitive disorders. The proposed multi-agent architecture is based on a five-layer system that autonomously develops high-level knowledge to detect anomalies in the home environment scenarios and therefore support the user. Bayesian networks (BNs) are exploited to perform high-level deductive reasoning on low-level multi-modal information, thus recognizing senseless or dangerous behaviors, environmental disruptions, changes in behavioral patterns, and serious medical events. Bi-directional user-system interaction provides user support by leveraging Speech-To-Text and Text-To-Speech AI agents. Three main functions were tested: real-time data integration, anomaly detection, and two-way interaction.

Publisher

Springer International Publishing

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3