Please use this identifier to cite or link to this item:
http://hdl.handle.net/1942/24553
Title: | learning algorithms for sensor interpretation on an exo-skeleton | Authors: | Bonné, Ruben | Advisors: | VANRUMSTE, Bart CUYPERS, Ludo GERGANOV, Todor |
Issue Date: | 2017 | Publisher: | UHasselt | Abstract: | COMmeto, active in software architecture services and software development, is involved together with 7 other partners in a European project called Axo-Suit to develop an assistive exo-skeleton for elderly people. COMmeto is responsible for the software architecture. In the case of the arm of the exo-skeleton the adjustment of the exo-skeleton to a person is carried out manually which takes a long time. This thesis focuses on the development of a machine learning algorithm to detect and classify the different movements in the sagittal plane and on the machine learning algorithms to adjust the exo-skeleton to the user. To develop the algorithm, the data used to adjust the exo-skeleton was defined to determine which movements are required. Additionally, the transfer of this data from the human intention detection (HID) to the workstation was solved. Furthermore, different algorithms were tested on the gathered data from different persons to carry out the tuning and classify the movements. Based on the results it can be stated that after a short learning period the various movements can be classified. To ensure the best results and to make the exo-skeleton arm as comfortable as possible, a different machine learning model is applied for each user. Another important factor to note is how well the sensor belts are attached to the user's arm. | Notes: | master in de industriële wetenschappen: elektronica-ICT | Document URI: | http://hdl.handle.net/1942/24553 | Category: | T2 | Type: | Theses and Dissertations |
Appears in Collections: | Master theses |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
00000000-64ef-426b-8e4c-2908061f16d7.pdf | 2.54 MB | Adobe PDF | View/Open | |
00000000-f9e1-48b8-9ec4-9daf7fa203e8.pdf | 1.19 MB | Adobe PDF | View/Open |
Page view(s)
18
checked on Sep 14, 2023
Download(s)
14
checked on Sep 14, 2023
Google ScholarTM
Check
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.