Abstract
This paper outlines research to improve robots’ ability to
ground their conversations with elderly people. We discuss a theoretical
framework on how to achieve this, and present a rough outline of how we
collect multimodal interactions between elderly and the QT robot with
the aim of using these data to fine-tune pre-trained models to detect
signals of miscommunication –initially from facial expressions and later
to be optimised using other modalities such as audio, body posture and
gesture.
ground their conversations with elderly people. We discuss a theoretical
framework on how to achieve this, and present a rough outline of how we
collect multimodal interactions between elderly and the QT robot with
the aim of using these data to fine-tune pre-trained models to detect
signals of miscommunication –initially from facial expressions and later
to be optimised using other modalities such as audio, body posture and
gesture.
| Original language | English |
|---|---|
| Number of pages | 4 |
| Publication status | Published - 2024 |
| Event | ASIMOV-2024: Workshop on Adaptive Social Interaction based on user’s Mental mOdels and behaVior in HRI - Odense, Denmark Duration: 23-Oct-2024 → 23-Oct-2024 https://easychair.org/cfp/asimov2024 |
Conference
| Conference | ASIMOV-2024 |
|---|---|
| Country/Territory | Denmark |
| Period | 23/10/2024 → 23/10/2024 |
| Internet address |
Fingerprint
Dive into the research topics of 'Towards grounding conversations between social robots and elderly persons'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver