DOLARS, a Distributed On-Line Activity Recognition System by Means of Heterogeneous Sensors in Real-Life Deployments-A Case Study in the Smart Lab of The University of Almería.

Javier Medina-Quero,Juan F Sanjuan,Pilar M Ortigosa,Marcos Lupión

doi:10.3390/s21020405

Javier Medina-Quero, Juan F Sanjuan + Show 2 more

Open Access

https://doi.org/10.3390/s21020405

Copy DOI

Abstract

Activity Recognition (AR) is an active research topic focused on detecting human actions and behaviours in smart environments. In this work, we present the on-line activity recognition platform DOLARS (Distributed On-line Activity Recognition System) where data from heterogeneous sensors are evaluated in real time, including binary, wearable and location sensors. Different descriptors and metrics from the heterogeneous sensor data are integrated in a common feature vector whose extraction is developed by a sliding window approach under real-time conditions. DOLARS provides a distributed architecture where: (i) stages for processing data in AR are deployed in distributed nodes, (ii) temporal cache modules compute metrics which aggregate sensor data for computing feature vectors in an efficient way; (iii) publish-subscribe models are integrated both to spread data from sensors and orchestrate the nodes (communication and replication) for computing AR and (iv) machine learning algorithms are used to classify and recognize the activities. A successful case study of daily activities recognition developed in the Smart Lab of The University of Almería (UAL) is presented in this paper. Results present an encouraging performance in recognition of sequences of activities and show the need for distributed architectures to achieve real time recognition.

Highlights

Ambient-assisted living has been adopted in several contexts, such as, smart homes and healthcare applications [4] aiming both at improving the quality of care services and allowing people to stay independent in their own homes for as long as possible [5]
We describe the case studies developed in the work to evaluate the performance of the approach in terms of computing time for on-line Activity Recognition (AR) and accuracy of AR approach based on sequences of daily human activities developed by an inhabitant
In addition it is noteworthy that all previous experiments have been carried out considering a single user in the smart home

Summary

Introduction

Smart environments allow the monitoring of human activity of inhabitants in an increasingly less invasive way [1]. Ambient-assisted living has been adopted in several contexts, such as, smart homes and healthcare applications [4] aiming both at improving the quality of care services and allowing people to stay independent in their own homes for as long as possible [5]. They aim to provide assistance to the inhabitant by, for instance, detecting emergency situations based on the data being observed [7]. In the initial work in the field of AR, binary ambient sensors [8] (such as passive infrarred sensors or open door) dominated the sensing of human behaviours being proposed as suitable devices for describing daily human activities. A new generation of devices has highlighted the capability of sensing the activity from a closer point of view of the user [9]

Methods

Results

Conclusion