WEKO3
アイテム
比例筋電位制御に向けた筋シナジーの抽出、解釈、および応用の研究
https://doi.org/10.18997/00007810
https://doi.org/10.18997/00007810081c4bb1-f4fc-4dff-913a-606cde39c70a
| 名前 / ファイル | ライセンス | アクション |
|---|---|---|
sei_k_372.pdf (3.8 MB)
|
|
| アイテムタイプ | 学位論文 = Thesis or Dissertation(1) | |||||||
|---|---|---|---|---|---|---|---|---|
| 公開日 | 2020-06-16 | |||||||
| 資源タイプ | ||||||||
| 資源タイプ識別子 | http://purl.org/coar/resource_type/c_db06 | |||||||
| 資源タイプ | doctoral thesis | |||||||
| タイトル | ||||||||
| タイトル | Extraction, Interpretation and Application of Nonlinear Muscle Synergies toward Proportional Myoelectric Control | |||||||
| 言語 | en | |||||||
| タイトル | ||||||||
| タイトル | 比例筋電位制御に向けた筋シナジーの抽出、解釈、および応用の研究 | |||||||
| 言語 | ja | |||||||
| 言語 | ||||||||
| 言語 | eng | |||||||
| 著者 |
Dwivedi, Sanjay Kumar
× Dwivedi, Sanjay Kumar
|
|||||||
| 抄録 | ||||||||
| 内容記述タイプ | Abstract | |||||||
| 内容記述 | Transfer of human intentions into myoelectric hand prostheses is generally achieved by learning a mapping, directly from sEMG signals to the Kinematics using linear or nonlinear regression approaches. Due to the highly random and nonlinear nature of sEMG signals such approaches are not able to exploit the functions of the modern pros- thesis, completely. Inspired from the muscle synergy hypothesis in the motor control community, some studies in the past have shown that better estimation accuracies can be achieved by learning a mapping to kinematics space from the synergistic features extracted from sEMG. However, mainly linear algorithms such as Principle Compo- nent Analysis （PCA）, and Non-negative matrix factorization （NNMF） were employed to extract synergistic features, separately, from EMG and kinematics data and have not considered the nonlinearity and the strong correlation that exist between finger kine- matics and muscles. To exploit the relationship between EMG and Finger Kinematics for myoelectric control, we propose the use of the Manifold Relevance Determination （MRD） model （multi-view learning） to find the correspondence between muscular and kinematics by learning a shared low-dimensional representation. In the first part of the study, we present the approach of multi-view learning, interpretation of extracted non- linear muscle synergies from the joint study of sEMG and finger kinematics and their use in estimating the finger kinematics for the upper-limb prosthesis. Applicability of the proposed approach is then demonstrated by comparing the kinematics estimation accuracies against linear synergies and direct mapping. In the second part of the study, we propose a new approach to extract nonlinear muscle synergies from sEMG using multiview learning which addresses the two main drawbacks （1. Inconsistent synergistic patterns upon addition of sEMG signals from more muscles, 2. Weak metric for accessing the quality and quantity of muscle synergies） of established algorithms and discuss the potential of the proposed approach for reducing the number of electrodes with negligible degradation in predicted kinematics. | |||||||
| 目次 | ||||||||
| 内容記述タイプ | TableOfContents | |||||||
| 内容記述 | 1 Introduction||2 Related Work||3 Extraction of nonlinear synergies for proportional and simultaneous estimation of finger kinematics||4 An Approach to Extract Nonlinear Muscle Synergies from sEMG through Multi-Model Learning||5 Conclusion and Future Work | |||||||
| 備考 | ||||||||
| 内容記述タイプ | Other | |||||||
| 内容記述 | 九州工業大学博士学位論文 学位記番号：生工博甲第372号 学位授与年月日：令和2年3月25日 | |||||||
| キーワード | ||||||||
| 主題Scheme | Other | |||||||
| 主題 | Electromyography (EMG) 筋電位信号 | |||||||
| キーワード | ||||||||
| 主題Scheme | Other | |||||||
| 主題 | multi-fingered hand多指ハンド | |||||||
| キーワード | ||||||||
| 主題Scheme | Other | |||||||
| 主題 | Nonlinear muscle and kinematic synergies運動学シナジー | |||||||
| キーワード | ||||||||
| 主題Scheme | Other | |||||||
| 主題 | dimensionality reduction次元削減 | |||||||
| アドバイザー | ||||||||
| 柴田, 智広 | ||||||||
| 学位授与番号 | ||||||||
| 学位授与番号 | 甲第372号 | |||||||
| 学位名 | ||||||||
| 学位名 | 博士（工学） | |||||||
| 学位授与年月日 | ||||||||
| 学位授与年月日 | 2020-03-25 | |||||||
| 学位授与機関 | ||||||||
| 学位授与機関識別子Scheme | kakenhi | |||||||
| 学位授与機関識別子 | 17104 | |||||||
| 学位授与機関名 | 九州工業大学 | |||||||
| 学位授与年度 | ||||||||
| 内容記述タイプ | Other | |||||||
| 内容記述 | 令和元年度 | |||||||
| 出版タイプ | ||||||||
| 出版タイプ | VoR | |||||||
| 出版タイプResource | http://purl.org/coar/version/c_970fb48d4fbd8a85 | |||||||
| アクセス権 | ||||||||
| アクセス権 | open access | |||||||
| アクセス権URI | http://purl.org/coar/access_right/c_abf2 | |||||||
| ID登録 | ||||||||
| ID登録 | 10.18997/00007810 | |||||||
| ID登録タイプ | JaLC | |||||||
