Tesi etd-05132022-182746 |
Link copiato negli appunti
Tipo di tesi
Tesi di laurea magistrale
Autore
MENOLOTTO, MARIANGELA
URN
etd-05132022-182746
Titolo
Combining optimal design and Kalman filtering for upper limb kinematic and EMG acquisition
Dipartimento
INGEGNERIA DELL'INFORMAZIONE
Corso di studi
INGEGNERIA ROBOTICA E DELL'AUTOMAZIONE
Relatori
relatore Prof. Bianchi, Matteo
correlatore Dott. Averta, Giuseppe Bruno
correlatore Bonifati, Paolo
correlatore Dott. Averta, Giuseppe Bruno
correlatore Bonifati, Paolo
Parole chiave
- EKF
- EMG
- fPCA
- IMU
- joints
- kinematic
- muscular
- MVE
- sensors
- upper limbs
Data inizio appello
01/06/2022
Consultabilità
Non consultabile
Data di rilascio
01/06/2025
Riassunto
The aim of the thesis was to develop a wearable systems for the evaluation of the kinematic and muscular state of the human upper limb through the use of the least number of sensors.
In order to achieve this goal, a collection of IMU and EMG data was recorded while the subjects of the experiment performed activities of daily living. An optimization procedure was carried out to find the optimal sensors to be measured and the remaining degrees of freedom were estimated through the use of a Minimum Variance Estimator.
Since the collected data were time-varying signals, a functional Principal Component Analysis was carried out on part of the dataset to extract the constant coefficients to be used in the MVE.
The performance of the estimator were validated computing the RMS error between the estimated and measured data.
In order to achieve this goal, a collection of IMU and EMG data was recorded while the subjects of the experiment performed activities of daily living. An optimization procedure was carried out to find the optimal sensors to be measured and the remaining degrees of freedom were estimated through the use of a Minimum Variance Estimator.
Since the collected data were time-varying signals, a functional Principal Component Analysis was carried out on part of the dataset to extract the constant coefficients to be used in the MVE.
The performance of the estimator were validated computing the RMS error between the estimated and measured data.
File
| Nome file | Dimensione |
|---|---|
La tesi non è consultabile. |
|