Título: System identification of a soft manipulator with polynomial methods
Autores: Tiago Barreto Sant’Anna, Pedro Machado, Emanuel Benício de Almeida Cajueiro & Lucas Cruz da Silva
Resumo: This paper explores system identification techniques within the domain of soft robotics, focusing specifically on a tendon-actuated soft manipulator. The primary goal is to perform system identification for this manipulator using two distinct mathematical models: ARX (Auto-Regressive with eXogenous input) and ARMAX (Auto-Regressive Moving Average with eXogenous input). The manipulator system is driven by three motors, and the piecewise constant curvature parameters, θ and ϕ, are considered as outputs. The comparative analysis reveals that the effectiveness of these identification methods varies according to system dynamics. The ARX model performs well with simpler dynamics like those of ϕ, while the ARMAX model proves superior for more complex or noisy systems like θ, owing to its inclusion of moving average components that capture additional dynamics. These findings underscore the importance of selecting the appropriate model based on specific system characteristics for accurate modeling and prediction.
Palavras-chave: Soft robots; system identification; polynomial methods; multivariable systems; soft manipulator.
Páginas: 7
Código DOI: 10.21528/CBIC2025-1173596
Artigo em PDF: CBIC_2025_paper1173596.pdf
Arquivo BibTeX:
CBIC_2025_1173596.bib