| 研究課題/領域番号 |
21K18105
|
|---|
| 研究種目 |
若手研究
|
| 配分区分 | 基金 |
|---|
| 審査区分 |
小区分90150:医療福祉工学関連
|
|---|
| 研究機関 | 九州大学 |
|---|
研究代表者 |
|
|---|
| 研究期間 (年度) |
2021-04-01 – 2024-03-31
|
| 研究課題ステータス |
交付 (2022年度)
|
| 配分額 *注記 |
4,680千円 (直接経費: 3,600千円、間接経費: 1,080千円)
2022年度: 1,300千円 (直接経費: 1,000千円、間接経費: 300千円)
2021年度: 3,380千円 (直接経費: 2,600千円、間接経費: 780千円)
|
|---|
| キーワード | HDEMG / Gabor filters / Motion Intention / HD EMG / High Density EMG |
|---|
| 研究開始時の研究の概要 |
Wearable robots need to be controlled according to the human motion intention. Existing technologies cannot predict the intentions intuitively. HDEMG contains information related to motor unit activations. This study investigates a new paradigm using HD EMG to estimate human motion intention.
|
| 研究実績の概要 |
One of the major challenge in HDEMG research is the electrode shift during muscle movement and between different experimental sessions. To address this we have being considering different techniques that can give enough information of variation of the muscle activity irrespective of the electrode's exact location. Thus, we have understood the spatial variation of the HDEMG signals can provide enough information to differentiate finger motions that have similar muscle activity pattern measure with SEMG signals. In the process we measured, HDEMG signals for 6 different types of finger motions. Then activations maps of the HDEMG signals were generated using the root mean square (EMS) values of the preprocessed HDEMG data. Gabor features were used extract spatial variations of the heat maps and error correcting output codes based multi class support vector machine classifier was used to classify 6 different finger motions with an average accuracy of 95.8%, in an offline study. The Gabor features were successful in extracting information related to the muscle activity from the heat maps. Further time series data of RMS values were used to train deep learning network to classify the 6 classes of motion with 85% accuracy.
|
| 現在までの達成度 (区分) |
現在までの達成度 (区分)
2: おおむね順調に進展している
理由
We have successfully estimated motion intention for 6 different types of closely related finger motions using spatial variations of the HDEMG activity both in an offline and in a realtime based study. Changing from the initial plan of using motor unit information identified from the HDEMG activations, spatial variations was a better feature in the current study to provide necessary information to differentiate finger motions. We are now designing different layout patterns that can successfully used to identify spatial variation of HDEMG activity for larger muscles for a wide range of estimation of motion intentions for upper limb motions.
|
| 今後の研究の推進方策 |
In the next step of the study we are are now planing to estimate motion intention for further closely related finger motions with a twelve different motions. In this study we are working on augmenting the available HDEMG data to provide more information related spatial variations of the muscle activity by using new techniques for feature extraction, in realtime.
At the same time, we will perform further studies using the fabricated sensor layouts to estimate motion intention of multi degree of freedoms of human upper limb motions.
|
