博士論文 - Romy Budhi Widodo - 九州工業大学(工学) - 拡張カルマンフィルタおよびニューラルネットを利用した靴型歩行計測装置の姿勢・歩幅推定に関する研究

書誌事項

タイトル: Research on Attitude Determination and Step-Length Prediction using Extended Kalman Filter and Artificial Neural Network for Shoe-Type Gait Measurement Device
タイトル別名: 拡張カルマンフィルタおよびニューラルネットを利用した靴型歩行計測装置の姿勢・歩幅推定に関する研究
著者名: Romy Budhi Widodo
学位授与大学: 九州工業大学 (大学ID:0071) (CAT機関ID:KI000844)
取得学位: 博士（工学）
学位授与番号: 甲生工第288号
学位授与年月日: 2017-03-24

注記・抄録

The information of the spatial and temporal gait parameters is useful for the physicians and physical therapists for determining a patient’s gait condition. For that purpose, we have developed a shoe-type measurement device to measure several gait parameters. This paper discusses the improvement of a shoe-type measurement device that consists of: 1) the elimination of the effect of external acceleration in attitude calculation; 2) the redesign of ultrasonic sensors positioning in order to predict step-length; and 3) information regarding the plantar center of pressure when people walk. The first improvement resulted from the sensors’ fusion, using the extended Kalman filter (EKF) with the proposed external acceleration compensation. The result of attitude determination shows an improvement in performance (mean square error) over the conventional EKF. The second improvement was made by using acoustic simulation and the implementation of the sensors’ positions to the actual shoes. The artificial neural network was used to accommodate the step-length prediction using two-layer architecture. The results of the predictive performance of step-length demonstrate an improvement in distance error compared to the previous study. The third improvement was made by redesigning the position of the pressure sensors, calibrating sensors, and calculating the center of pressure using tactile force sensors.

九州工業大学博士学位論文学位記番号:生工博甲第288号学位授与年月日:平成29年3月24日

1. Introduction|2. Previous Studies|3. Attitude Sensing using a Kalman Filtering Based Algorithm to Compensate for External|4. Ultrasonic Sensors Position and Step-Length Prediction using Artificial Neural Network in a Shoe-Type Gait Measurement Device|5. Pressure Sensing|6. Conclusions and Future Direction

平成28年度

九州工業大学博士学位論文（要旨）学位記番号：生工博甲第288号学位授与年月日：平成29年3月24日

キーワード

step-length prediction Extended Kalman filter shoe-type measurement device Center of Pressure external acceleration 歩幅予測拡張カルマンフィルタ靴型計測装置圧力中心外部加速度

各種コード

NII論文ID(NAID)

500001036693

NII著者ID(NRID)

8000001129915

本文言語コード

データ提供元

機関リポジトリ (国立大学法人九州工業大学)
NDLデジタルコレクション

外部リンク - 機関リポジトリを含む

九州工業大学(工学) - 博士論文

九州工業大学(工学) - 博士論文をもっと見る

工学 - 博士論文

工学 - 博士論文をもっと見る