A Framework Towards Ambient Assisted Living Enhanced by Service Robots

Abstract

Automated Human Activity Recognition (HAR) is a crucial technology for Ambient Assisted Living applications (AAL), which uses sensor data to detect and recognize human activities and behaviors. However, this technology faces several challenges, such as coping with the inherent variability of human motions, the presence of multiple people, nuisance factors such as partial occlusion, changes in lighting or viewpoint, etc. The proposed research work of ACTIVE aims to address these challenges by developing a digital assistant capable of accurately recognizing a wide range of human activities by introducing new methods and technologies for dynamic, adaptive multimodal data fusion and analysis. Additionally, the research work focuses on the incorporation of a mobile robotic platform to assist in the monitoring of human activities in dynamic environments. The ultimate goal is to provide a better understanding of human behavior, improve quality of life, and enable personalized services for individuals.