{"id":{"repo_id":"uwtsd","oai_identifier":"oai:repository.uwtsd.ac.uk:2363"},"canonical_url":"https://search.dev.ndltd.org/etd/uwtsd/oai:repository.uwtsd.ac.uk:2363","repository":{"repo_id":"uwtsd","name":"University of Wales Trinity Saint David","base_url":"https://repository.uwtsd.ac.uk/cgi/oai2"},"display":{"title":"A Study of Machine Learning based on Physical Load Classification","abstract":"A whole system of gait data acquisition based on IMU sensors was completed in this study and an in-depth study was carried out based on the data collected by this system. The IMU data of human lower limb movements were experimentally verified to be usable for physical load recognition and classification work. As a result, the processed data used for machine learning is divided into three main categories: no-load, light-load and heavy-load. The traditional LSTM was able to achieve 71.1% accuracy for the multi-classification problem, and the Bi-LSTM was able to achieve a 74.5% correct classification rate. The excellent performance of multiple Bi-LSTMs for binary classification problems was exploited by changing the discriminative approach of the network, and a correct recognition rate of 94.1% was achieved for 3 classifications using 3 binary Bi-LSTM networks. This method can be used to provide a reference of the drive torque size and drive mode for the drive assist unit of an exoskeleton robot or a sports rehabilitation machine to achieve automatic determination of the torque size of the assist. It can reduce the energy consumption of the wearer during walking, reduce the fatigue during weight-bearing, increase the power-assist efficiency of the system and improve the performance of the system, which is of good reference value.","abstract_html":"A whole system of gait data acquisition based on IMU sensors was completed in this study and an in-depth study was carried out based on the data collected by this system. The IMU data of human lower limb movements were experimentally verified to be usable for physical load recognition and classification work. As a result, the processed data used for machine learning is divided into three main categories: no-load, light-load and heavy-load. The traditional LSTM was able to achieve 71.1% accuracy for the multi-classification problem, and the Bi-LSTM was able to achieve a 74.5% correct classification rate. The excellent performance of multiple Bi-LSTMs for binary classification problems was exploited by changing the discriminative approach of the network, and a correct recognition rate of 94.1% was achieved for 3 classifications using 3 binary Bi-LSTM networks. This method can be used to provide a reference of the drive torque size and drive mode for the drive assist unit of an exoskeleton robot or a sports rehabilitation machine to achieve automatic determination of the torque size of the assist. It can reduce the energy consumption of the wearer during walking, reduce the fatigue during weight-bearing, increase the power-assist efficiency of the system and improve the performance of the system, which is of good reference value.","abstract_has_math":false,"creators":["Zhao, Yuxuan"],"institution":"University of Wales Trinity Saint David","degree_name":"msc","degree_level":"masters","degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2022,"date_issued":"2022-12","date_published":"2022-12","updated_at":"2026-07-24T05:52:56Z","subjects":["QA75 Cyfrifiaduron electronig"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier.grantnumber","label":"Dc Identifier Grantnumber","values":["UWTSD"],"render_values":[{"text":"UWTSD","href":null,"code":true}]}]},"links":{"outbound_url":"https://doi.org/10.82227/repository.uwtsd.ac.uk.00002363","outbound_label":"DOI","outbound_source":"dc:identifier.doi"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.sponsor","label":"Sponsor","values":["University of Wales Trinity Saint David"]},{"key":"dc:creator","label":"Author","values":["Zhao, Yuxuan"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2022-12-07"]},{"key":"dc:date.issued","label":"Date","values":["2022-12"]},{"key":"dc:publisher.commercial","label":"Dc Publisher Commercial","values":["University of Wales Trinity Saint David"]},{"key":"dc:publisher.department","label":"Dc Publisher Department","values":["Traethodau Meistr","Wales Institute for Science and Art: Applied Computing"]},{"key":"dc:publisher.institution","label":"Dc Publisher Institution","values":["University of Wales Trinity Saint David"]},{"key":"dc:relation.isreferencedby","label":"Dc Relation Isreferencedby","values":["https://repository.uwtsd.ac.uk/id/eprint/2363/"]},{"key":"dc:type","label":"Dc Type","values":["Gosodiad"]},{"key":"dc:type.qualificationlevel","label":"Dc Type Qualificationlevel","values":["masters"]},{"key":"dc:type.qualificationname","label":"Dc Type Qualificationname","values":["msc"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["QA75 Cyfrifiaduron electronig"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.doi","label":"DOI","values":["10.82227/repository.uwtsd.ac.uk.00002363"]},{"key":"dc:identifier.grantnumber","label":"Dc Identifier Grantnumber","values":["UWTSD"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://repository.uwtsd.ac.uk/id/eprint/2363/1/Zhao%2C%20Yuxuan%20%282022%29%20MSc%20A%20study%20of%20machine%20learning%20based%20on%20physical%20load%20classification.pdf"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["A whole system of gait data acquisition based on IMU sensors was completed in this study and an in-depth study was carried out based on the data collected by this system. The IMU data of human lower limb movements were experimentally verified to be usable for physical load recognition and classification work. As a result, the processed data used for machine learning is divided into three main categories: no-load, light-load and heavy-load. The traditional LSTM was able to achieve 71.1% accuracy for the multi-classification problem, and the Bi-LSTM was able to achieve a 74.5% correct classification rate. The excellent performance of multiple Bi-LSTMs for binary classification problems was exploited by changing the discriminative approach of the network, and a correct recognition rate of 94.1% was achieved for 3 classifications using 3 binary Bi-LSTM networks. This method can be used to provide a reference of the drive torque size and drive mode for the drive assist unit of an exoskeleton robot or a sports rehabilitation machine to achieve automatic determination of the torque size of the assist. It can reduce the energy consumption of the wearer during walking, reduce the fatigue during weight-bearing, increase the power-assist efficiency of the system and improve the performance of the system, which is of good reference value."]},{"key":"dc:format","label":"Dc Format","values":["text"]},{"key":"dc:title","label":"Title","values":["A Study of Machine Learning based on Physical Load Classification"]}]}],"canonical_facts":{"dc:contributor.sponsor":["University of Wales Trinity Saint David"],"dc:creator":["Zhao, Yuxuan"],"dc:date":["2022-12-07"],"dc:date.issued":["2022-12"],"dc:description.abstract":["A whole system of gait data acquisition based on IMU sensors was completed in this study and an in-depth study was carried out based on the data collected by this system. The IMU data of human lower limb movements were experimentally verified to be usable for physical load recognition and classification work. As a result, the processed data used for machine learning is divided into three main categories: no-load, light-load and heavy-load. The traditional LSTM was able to achieve 71.1% accuracy for the multi-classification problem, and the Bi-LSTM was able to achieve a 74.5% correct classification rate. The excellent performance of multiple Bi-LSTMs for binary classification problems was exploited by changing the discriminative approach of the network, and a correct recognition rate of 94.1% was achieved for 3 classifications using 3 binary Bi-LSTM networks. This method can be used to provide a reference of the drive torque size and drive mode for the drive assist unit of an exoskeleton robot or a sports rehabilitation machine to achieve automatic determination of the torque size of the assist. It can reduce the energy consumption of the wearer during walking, reduce the fatigue during weight-bearing, increase the power-assist efficiency of the system and improve the performance of the system, which is of good reference value."],"dc:format":["text"],"dc:identifier.doi":["10.82227/repository.uwtsd.ac.uk.00002363"],"dc:identifier.grantnumber":["UWTSD"],"dc:identifier.uri":["https://repository.uwtsd.ac.uk/id/eprint/2363/1/Zhao%2C%20Yuxuan%20%282022%29%20MSc%20A%20study%20of%20machine%20learning%20based%20on%20physical%20load%20classification.pdf"],"dc:publisher.commercial":["University of Wales Trinity Saint David"],"dc:publisher.department":["Traethodau Meistr","Wales Institute for Science and Art: Applied Computing"],"dc:publisher.institution":["University of Wales Trinity Saint David"],"dc:relation.isreferencedby":["https://repository.uwtsd.ac.uk/id/eprint/2363/"],"dc:subject":["QA75 Cyfrifiaduron electronig"],"dc:title":["A Study of Machine Learning based on Physical Load Classification"],"dc:type":["Gosodiad"],"dc:type.qualificationlevel":["masters"],"dc:type.qualificationname":["msc"]},"updated_at":"2026-07-24T05:52:56Z"}