Leveraging Digital Twins and Generative AI to Alleviate Loneliness Among Elderly Adults Living Alone Through Smart Flowerpot Design

Loneliness and Social Isolation from elder people present a critical factor in different diseases of aging, necessitating tailored intervention for individuals to effectively engaging with interventions and interacting to their stakeholders. A great deal of research efforts has been made to tackle the challenging issues related to loneliness and social isolations of elder people, mainly focusing on develop various personal and group interventions such as social robots, group/community activity interventions, serious games, and Chatbots. The key barrier to have long-term impacts from these interventions is to make interventions adaptable to individual’s personal needs, scenarios/environments and contexts. To overcome the barrier, how to design adaptable interventions and implement interventions adaptably to meet a person’s needs is a great research question. This research aims to address this research question by (1) applying adaptable design principles into smart intervention design of a smart flowerpot or pet plant, enabled by generative AI and reconfigurable design, (2) applying digital twin technology to create a user’s personal replica with only needed information to support personalized intervention adaption. The objective is to create a smart service system/platform based on Digital Twins and Generative AI to effectively design and deliver smart interventions, for and with elderly individuals and their stakeholders in a smart service ecosystem way. The smart flowerpot is purposely designed to make it like a normal pot flower like most families having them at home, its visitor/user can trim, water, move and care it. The home users typically visit their home flowerpots regularly if not daily. And if a flowerpot becomes very attractive to its users, it can become a pet plant befriending with its users. Some reports already indicates that pet plant can be a useful means for companying elder people at home and make flowerpot-based intervention easier to adapt. On the other hand, to turn an ordinary home flowerpot into a smart flowerpot, we design a special structure for a flowerpot, which not only support a normal flower plant to grew but also can smartly interact with its users/visitors/cares. Within the structure, we equip some IoT sensors to sense a visitor presence, and then connect to the visitor’s personal digital twin to gain better understanding of the visitor’s needs, and finally feed the user’s needs into generative AI and reconfigurable designed elements to provide personalised interventions such as coloured light, display patterns, background music and chats generated by generative AI, etc. This paper presents our research processes, methods and evaluations via a smart flowerpot design, development and testing. The qualitative evaluation shows that our proposed adaptable intervention design and implementation system enabled by digital twin and generative AI technologies has a great potential of bettering understanding of a user’s needs, user engagement with interventions and user experience. Thus, this research has the potential to herald a new era in healthcare for the aging population, fostering smooth smart service and improved quality of life for older adults.