Multi-Agent Interaction to Assist Visually-Impaired and Elderly People

Area of Science:

• Computer Science
• Artificial Intelligence
• Human-Computer Interaction

Background:

• Older adults face unique challenges like cognitive decline and solitude.
• Voice-controlled smart home systems offer potential solutions for proactive support.
• Multi-Agent Systems (MAS) are suitable for dynamic, distributed environments like Ambient Intelligence (AmI).

Purpose of the Study:

• To adapt a MAS-based voice-controlled smart home system for the elderly.
• To evaluate the system's suitability for addressing the needs of older adults.
• To identify areas for improvement in human-agent interaction within AmI.

Main Methods:

• Utilized a Multi-Agent System (MAS) platform.
• Applied a four-step interactive design process.
• Adapted functionalities previously developed for visually impaired users.

Main Results:

• The adapted system demonstrated elements conducive to natural user interaction.
• The approach showed potential for supporting older adults with cognitive decline and solitude.
• Identified specific improvements and new features for future AmI system development.

Conclusions:

• The MAS-based voice-controlled system offers a viable approach for supporting older adults.
• Natural interaction is key for effective Ambient Intelligence (AmI) applications.
• Further development can enhance AmI systems for greater user adaptability and support.