Daily living activity recognition with ECHONET Lite appliances and motion sensors

Recently, IoT (Internet of Things) technologies have been attracting increasing attention. Among many applications of IoT, homes can be the most promising target. One of the purposes to deploy IoT in homes is automatic recognition of activities of daily living (ADLs). It is expected that ADL recognition in homes enables many new services such as elderly people monitoring and low energy appliance control. In existing studies on ADL recognition, however, it is hard to build a system to acquire data for ADL recognition in terms of installation cost. In this paper, we propose a method that reduces costs of the ADL recognition system by using ECHONET Lite-ready appliances which are expected to be widely spread in the future. ECHONET Lite is a communication protocol for control and sensor networks in smart-homes and standardized as ISO/IEC-4-3. The proposed method utilizes information (e.g., on/off state) from appliances and motion sensors attached to them as features and recognizes ADLs through machine learning. To evaluate the proposed method, we collected data in our smart-home testbed while several participants are living there. As a result, the proposed method achieved about 68% classification accuracy for 9 different activities.

[1]  Walid Bourennane,et al.  Activity monitoring system for elderly in a context of smart home , 2013 .

[2]  Nikolaos Papanikolopoulos,et al.  Multi-Camera Human Activity Monitoring , 2008, J. Intell. Robotic Syst..

[3]  Kenji Mase,et al.  Activity and Location Recognition Using Wearable Sensors , 2002, IEEE Pervasive Comput..

[4]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[5]  Yutaka Arakawa,et al.  Exploring Accuracy-Cost Tradeoff in In-Home Living Activity Recognition Based on Power Consumptions and User Positions , 2015, 2015 IEEE International Conference on Computer and Information Technology; Ubiquitous Computing and Communications; Dependable, Autonomic and Secure Computing; Pervasive Intelligence and Computing.

[6]  Miguel A. Labrador,et al.  A Survey on Human Activity Recognition using Wearable Sensors , 2013, IEEE Communications Surveys & Tutorials.

[7]  Hongnian Yu,et al.  Elderly activities recognition and classification for applications in assisted living , 2013, Expert Syst. Appl..

[8]  John Krumm,et al.  PreHeat: controlling home heating using occupancy prediction , 2011, UbiComp '11.

[9]  Hirozumi Yamaguchi,et al.  Survey of Real-time Processing Technologies of IoT Data Streams , 2016, J. Inf. Process..

[10]  Prashant J. Shenoy,et al.  SmartCap: Flattening peak electricity demand in smart homes , 2012, 2012 IEEE International Conference on Pervasive Computing and Communications.

[11]  Lei He,et al.  Human activity recognition based on feature selection in smart home using back-propagation algorithm. , 2014, ISA transactions.

[12]  Chris D. Nugent,et al.  A Knowledge-Driven Approach to Activity Recognition in Smart Homes , 2012, IEEE Transactions on Knowledge and Data Engineering.

[13]  Jesse Hoey,et al.  Value-Directed Human Behavior Analysis from Video Using Partially Observable Markov Decision Processes , 2007, IEEE Transactions on Pattern Analysis and Machine Intelligence.