「世大智科/天才家居」-我們創業囉
Contact Professor: Yeh-Liang Hsu (徐業良)

103學年度元智大學機械工程研究所姜依帆碩士論文

Master thesis by Yi-FanJiang, Mechanical Engineering Department, Yuan Ze University, 2015

103碩士論文:以加速度感測器為核心發展穿戴式裝置個人生活型態模式分析技術

穿戴式裝置逐漸成為生活中的新潮流,其中三軸加速度感測器通常用於穿戴式裝置來測量用戶的動作信號,較無測量位置限制,亦不需電極緊貼皮膚,目前市售產品常以三軸加速度感測器所量測的步數為基礎,並搭配使用者身高和體重計算行走距離和能量消耗等數據,然而僅以步數做為估計往往運動強度或能量消耗無法正確被辨識。 本研究目的在利用穿戴式裝置加速度感測器所得到的使用者生理數值,進行活動力指標與行為模式相似度分析,並發展個人生活型態模式分析技術,以三軸加速度感測器中加速度、步頻兩種穿戴式裝置經常輸出的生理數據做為運動樣本,利用機器學習軟體Weka進行運動等級分類訓練,並以每日總能量消耗(Total Daily Energy Expenditure, TDEE)和基礎代謝率(Basal metabolic rate, BMR)進行單日PAL(Physical Activity Level)推估,轉換成對使用者更具實質意義的個人生活型態模式資訊。 本研究以低功耗藍芽(BLE)技術為基礎發展失智症患者防走失系統,並結合個人生活型態分析與區域定位功能,作為失智症患者應用案例,令穿戴式裝置發展帶來更多元服務。 關鍵字:生理數據、機器學習、低功耗藍芽、區域定位、防走失系統

Analysis of Personal Life Patterns Using Accelerometer Based Wearable Devices

Wearable devices are becoming increasingly popular. Many commercially available wearable devices are equipped with sensors to measure motion and physiological signals of the users. G-sensor is commonly used in such wearable devices to measure motion signals of the user. It is less restricted in measuring positions and there is no need to have electrodes touching the skin. Therefore, it is more convenient and flexible to design a wearable device with G-sensor, and it is often the only sensor in wearable devices for counting steps and detecting sleep duration. Such wearable devices often work with mobile device applications (Apps) for further data processing and display. Algorithms based on step count, height and weight of the user are used in the Apps for estimating travel distance and energy expenditure. However, the estimation based on step count is often pretty rough. In other words, intensities of physical activities (and therefore energy expenditure) cannot be correctly classified based on step count alone. The purpose of this study is to interpret users’ motion signals measured from G-sensor into physical activity intensity, then to calculate the activity index, similarity analysis of behavior patterns. And two features derived from the motion signals sensed by the G-sensor, average cadence and ratio of high G value, were used to classify the physical activities into four intensity levels. Eighty physical activity samples were collected and trained by machine learning software Weka to form a classification model, which can predict physical activity intensity from these two features. From the physical activity intensity, total daily energy expenditure (TDEE), physical activity levels (PAL) and personal life patterns can be derived. Further services can then be tailored for the wearable device user based on the data provided by wearable devices. Many older adults were requested to wear GPS based wearable devices or RFID tags, mainly for positioning or locali
Academic Thesis
Presentation Files
Last Updated:2016/7/21