Development of Brain Computer Interface (BCI) based control algorithm for master-slave configuration /
Aim of this project is to apply algorithms to classify and spectral estimation of arm and wrist generated EEG signals. Accuracy of angular movement depending on brain generated signal is very poor. In order to achieve higher accuracy multiple imaginary body part movement used. This project used a da...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
Kuala Lumpur :
Kulliyyah of Engineering, International Islamic University Malaysia,
2018
|
| Subjects: | |
| Online Access: | Click here to view 1st 24 pages of the thesis. Members can view fulltext at the specified PCs in the library. |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Aim of this project is to apply algorithms to classify and spectral estimation of arm and wrist generated EEG signals. Accuracy of angular movement depending on brain generated signal is very poor. In order to achieve higher accuracy multiple imaginary body part movement used. This project used a database created by The Berlin BCI group for the BCI Competition. The data of this database was collected from motor imagery of their right or left arm and wrist movement and EEG (Electroencephalogram) signal reading was being recorded from scalp by using electrodes. Further data collected from IIUM Bio-mechatronics lab using EEG headset, in order to verify with real human data. The Linear Discriminant Analysis (LDA) classifier entries were generated from three different frequencies in μ-rhythm (8 a I3Hz) and ~-rhythm (14 a 25Hz) by using the Windowed Means Paradigm method. In This experiment we train up the system by calibration data and then cross check the random evaluation data with the trained model. For the data collected from online, error rate was less than 0.17 %. The lab generated data did not give proper result because of high noise to signal ratio. Further research necessary to identify artefacts accurately. Based on these algorithms using Brain-Computer Interface (BCI), it is possible to control devices angular movement using imagery hand and wrist movement. To demonstrate that we assemble a microcontroller based robot. The research objectives have been fulfilled via simulation and experimental validation through hardware implementation. The experimental results show effective and reliable results and the novelty lies in the use of simple technique to achieve the objectives. As a whole, this work has investigated and analysed several classification techniques and adopted the best classifier to control master-slave configuration. |
|---|---|
| Physical Description: | xvi, 118 leaves : illustrations ; 30cm. |
| Bibliography: | Includes bibliographical references (leaves 95-101). |