Development of a Sensor-Based Ecosystem for Measuring Comfort and Activities in a Multi-Resident Context: the Age-SenseAI Project

The aging population is rapidly growing, increasing the demand for innovative solutions to support elderly individuals while minimizing the burden on caregivers. This paper presents the Age-SenseAI project, a novel measurement ecosystem designed to monitor comfort and activities in multi-resident environments. The system integrates a network of non-invasive environmental and physiological sensors, combined with Artificial Intelligence (AI) and data fusion techniques, to assess daily activities, indoor comfort, and potential health risks. A co-design approach involving professionals was adopted to define technical requirements, ensuring compliance and ethical considerations. The proposed sensor network collects real-time data, enabling personalized comfort assessments and detection of behaviour. Two primary use cases were developed: activity recognition in multi-resident contexts and indoor comfort assessment, integrating both objective environmental parameters and subjective user feedback. The architecture leverages cloud-based processing and AI-driven analytics to provide real-time insights and adaptive control mechanisms, enhancing elderly autonomy and safety. Future research will focus on improving personalization, deep learning models, and validating the ecosystem in real-world multi-resident scenarios. The Age-SenseAI project represents a significant step toward scalable, intelligent monitoring solutions for elderly care.