Upravljanje robotom pomoću električnih impulsa mišića

master's thesis

• Title (english): Robot control by measuring muscle electrical impulses
• Author: Lucija Marković
• Mentor: Bojan Jerbić (mentor)
• Granter: University of Zagreb; Faculty of Mechanical Engineering and Naval Architecture; Zagreb
• Defense date and country: 2019-07-11, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Mechanical Engineering

Abstract

Istraživanja na području interakcije robota i čovjeka od velike su važnosti u današnje vrijeme. Cilj tog područja je izrada sustava koji će se uklopiti u čovjekovo okruženje i raditi na čovjeku prirodniji način te uspostavljanje kvalitetnog međudjelovanja čovjeka i robotskog sustava. U ovom radu razrađena je ideja korištenja elektromiografskog signala kao informacije o silama u mišićima ruke. Time bi se robotskom sustavu dao uvid u sile kojima čovjek djeluje na okolinu. U radu je razjašnjena teorija elektromiografije i neuronske mreže, algoritma korištenog za estimaciju sile. Zatim je izrađen sustav koji estimira silu dvoglavog nadlaktičnog mišića prema podacima prikupljenima na više ispitanika. Nakon toga su analizirane značajke signala kako bi se dobio uvid u razlog velikih odstupanja dobivenih rezultata od željenih. Na temelju analize donesen je zaključak da je potrebno razraditi drugačiji model. Napravljen je sustav prema podacima samo jednog ispitanika te obrađena analiza s podacima prikupljenima u više mjerenja. U korisničkom sučelju implementiran je sustav u realnom vremenu. Na kraju je proučeno korištenje sustava na ostalim mišićima nadlaktice i podlaktice.

Abstract (english)

Nowadays, research in the field of Human Robot Interaction is of great importance. The goal of this field is production of effective interaction and creation of systems which would fit into the human environment and work in a more natural way. This paper presents the idea of using electromyographic signal as information on the force developed in the muscle. That would give the robot system an insight into the forces by which human affects the environment. This paper deals with the theory of electromyography and artificial neural network, the algorithm used for force estimation. Firstly, a system for force estimation of biceps brachii was developed using data from multiple subjects. Afterwards, the signal features were analyzed to gain insight into the reason for large deviations of the obtained results in comparison to the ones that are desired. According to conclusions from the analysis, another system was developed using a different approach. The system was made using data from only one subject and the data from different measurements were analyzed. This system was implemented in real time in the user interface. Finally, the use of the system on the other muscles of the upper arm and the forearm was researched.

• Keywords: mišić; sila; elektromiografija; neuronska mreža; Arduino; MATLAB
• Keywords (english): muscle; force; electromyiography; neural network; Arduino; MATLAB
• Language: croatian
• URN:NBN: urn:nbn:hr:235:879065
• Study programme: Title: Mechanical Engineering; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka strojarstva (magistar/magistra inženjer/inženjerka strojarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2019-07-04 08:33:49