On Using Smart Sensor Technologies for Indoor Activities Tracking and Localization

利用智能傳感器技術進行室內活動跟蹤和定位

Student thesis: Doctoral Thesis

View graph of relations

Related Research Unit(s)

Detail(s)

Awarding Institution
Supervisors/Advisors
Award date9 Apr 2019

Abstract

With rapid advances in sensor and wireless communication technologies, the research on using smart sensors for tracking patients with dementia in their daily living has received tremendous attentions in recent years. In this thesis, we studied how to use various types of smart sensors, e.g., 3D-depth camera, motion sensors, accelerometers, gyroscopes and magnetometers, to design and develop effective systems to track the daily activities of the patients with dementia within their living rooms. The tracked activity data can be used as important indicators of their health and long term statistical health data can describe their important health trend over the time. In order to maintain a healthy living both physiologically and psychologically, it is important for the patients to maintain an active daily life.

In this thesis, we first introduced our indoor activity tracking system called SmartMind. The main sensor adopted in SmartMind is MS Kinect which captures the images and generate 3D posture data of the user within his/her living room. The tracked activities are recorded in an activity database. After generating the activity records, various statistics can be generated to describe the daily living patterns of the user for predicting his/her current health trends. These records can also be reviewed by the authorised patient's relatives and medical professionals, e.g., community nurses, to assess his/her self-caring abilities. In addition to activity detection, SmartMind can also detect falls which can be a serious risk to the user.

Although SmartMind is effective in activity tracking, it has the privacy concern since most of the patients do not like to be monitored by cameras continuously. To resolve this limitation of SmartMind, we have explored the use of simple kinetic and magnetic sensors for activity tracking with the development of the system called ActiveLife. In ActiveLife, we have demonstrated how to use simple smart sensors, e.g., accelerometers, gyroscopes and magnetometers, to design and develop a system for effective tracking of the current activities being preformed by the user within his/her living rooms. In order to simplify the activity detection process, in ActiveLife, we adopt the context-based approach to model the common activities performed by the user in a day. Since the accelerometer and gyroscope are tri-axial sensors, the sensor data for different axes can be used to predict the current posture of the user while he/she is performing an activity. Combining with the heading direction of the posture obtained from the magnetometer and distance traveled during the transition of activities, we can estimate the current location of the user within the living rooms as well as the activity performing by the user. To further improve the estimation accuracy, we have designed an algorithm using the machine-learning technique, i.e., support vector machines (SVM), for activity classification.

Although it is shown that SmartMind and ActiveLife can track the activities of the user using simple sensors, sometime, it is necessary that the we may wish to track the location of the user in a more accurate result. In order to obtain accurate location information of the user, in the last part of the thesis, we will discuss our works on indoor localization using pervasive computing technologies. By using the pervasive device such as a smartphone, attached with the user, we have designed a system using the WLAN and RFID to locate the user indoor by comparing the RSSI from different base station.

    Research areas

  • Health Informatics, Mild Cognitive Impairment, Pervasive Computing, Motion Detection, Context-Aware Computing, Assistive Technology