EMG as a control parameter in lower limb prosthetics : surface vs. implanted electrodes

Human locomotion is controlled by complex neurological patterns. These patterns activate and synchronise the musculoskeletal system to enable efficient and graceful ambulation. A limb amputation results in a partial loss of the complex structure needed for human locomotion. Various prosthetic devices have been created as substitutes, however they have simply been unable to match the human nervous systems ability to move and orient the body in a smooth, efficient and intuitive manner. It is therefore logical to utilize the existing control of the nervous system and capitalise upon its enormous ability. Myoelectric sensors are one method of achieving that by providing the prosthesis with information about the amputees’ intentions through voluntary and involuntary control. The feasibility of using myoelectric signals, recorded with surface and implanted myoelectric sensors, for the control of a lower limb prosthesis was examined in this thesis. The pros and cons of the two recording methods were discussed and their potential to provide a viable control signals was examined. Additionally, pattern recognition algorithms were implemented to determine various phases during walking, stair descent and sitting/standing. The results show that a viable control signal can be recorded with the sensors. Furthermore, the signals could be used to increase the classification accuracy of existing prosthetic control schemes. However, the quality and robustness of the signals recorded with the implanted sensors far exceed those of the surface signals. In conclusion myoelectric control has a significant role in the next generation of lower limb prosthetics despite the challenges discussed in this thesis.

Author dc:creator

• Jóna Sigurðardóttir 1990-

Contributors dc:contributor

• Háskólinn í Reykjavík

• Heilbrigðisverkfræði
• Rafstraumur
• Gervilimir
• Gervigreind
• Tækni- og verkfræðideild
• Meistaraprófsritgerðir
• Biomedical engineering
• Myoelectric control
• Prosthesis
• Artificial intelligence
• School of Science and Engineering

Language dc:language.iso

en

Handle dc:identifier.uri

http://hdl.handle.net/1946/22318

OAI identifier oai:identifier

oai:skemman.is:1946/22318