Título:	Dynamic reconfiguration for multi-magnet tracking in myokinetic prosthetic interfaces
Autor(es):	Mendez, Sergio Andres Pertuz Mendes, Davi de Alencar Gherardini, Marta Arboleda, Daniel Marinho Muñoz Ayala, Helon Vicente Hultmann Cipriani, Christian
ORCID:	https://orcid.org/0000-0002-6311-3251 https://orcid.org/0000-0002-2156-5516 https://orcid.org/0000-0002-9219-6463 https://orcid.org/0000-0001-5406-3902 https://orcid.org/0000-0003-2108-0700
Afiliação do autor:	TU Dresden, Institute of Computer Engineering University of Brasilia, Department of Mechanical Engineering Scuola Superiore Sant’Anna, BioRobotics Institute University of Brasilia, Department of Mechanical Engineering, Electronics Engineering Undergraduate Program Pontifical Catholic University of Paraná, Department of Mechanical Engineering Scuola Superiore Sant’Anna, BioRobotics Institute
Assunto:	Prótese Aprendizado de máquina Reconfiguração dinâmica parcial Interface miocinética Field Programmable Gate Arrays (FPGAs)
Data de publicação:	19-Set-2024
Editora:	IEEE
Referência:	MENDES, Sergio Andres Pertuz et al. Dynamic reconfiguration for multi-magnet tracking in myokinetic prosthetic interfaces. IEEE transactions on medical robotics and bionics, v. 6, n. 4, p. 1678-1687, 2024. DOI: 10.1109/TMRB.2024.3464093. Disponível em: https://ieeexplore.ieee.org/document/10684318. Acesso em: 25 ago. 2025.
Abstract:	Recently myokinetic interfaces have been proposed to exploit magnet tracking for controlling bionic prostheses. This interface derives information about muscle contractions from permanent magnets implanted into the amputee’s forearm muscles. Machine learning models have been mapped on Field Programmable Gate Arrays (FPGAs) to track a single magnet, achieving good precision and computational efficiency, but consuming a large area and hardware resources. To track several magnets, here we propose a novel solution based on dynamic partial reconfiguration, switching three prediction models: a linear regressor, a radial basis function neural network, and a multi-layer perceptron neural network. A system with five magnets and 128 magnetic sensor inputs was used and experimental data were collected to train a system with five hardware predictors. To reduce the complexity of the models, we applied principal component analysis, ranking by correlation the number of inputs of each model. This run-time reconfigurable solution allows the circuits to be reconfigured in order to select the most reliable predictor model for each magnet while the rest of the circuit continues to operate extracting the most significant information from the captured signals. Thus, the proposed solution remarkably reduces the hardware occupation and improves the computational efficiency compared to previous solutions.
Unidade Acadêmica:	Faculdade de Tecnologia (FT) Departamento de Engenharia Mecânica (FT ENM)
Programa de pós-graduação:	Programa de Pós-Graduação em Sistemas Mecatrônicos
DOI:	10.1109/TMRB.2024.3464093
Versão da editora:	https://ieeexplore.ieee.org/document/10684318
Aparece nas coleções:	Artigos publicados em periódicos e afins

Os itens no repositório estão protegidos por copyright, com todos os direitos reservados, salvo quando é indicado o contrário.