Magneto-Inertial Measurement Units (MIMUs) are a valid alternative tool to optical stereophotogrammetry in human motion analysis. The orientation of a MIMU may be estimated by using sensor fusion algorithms. Such algorithms require input parameters that are usually set using a trial-and-error (or grid-search ) approach to find the optimal values. However, using trial-and-error requires a known reference orientation, a circumstance rarely occurring in real-life applications. In this article, we present a way to suboptimally set input parameters, by exploiting the assumption that two MIMUs rigidly connected are expected to show no orientation difference during motion. This approach was validated by applying it to the popular complementary filter by Madgwick et al. and tested on 18 experimental conditions including three commercial products, three angular rates, and two dimensionality motion conditions. Two main findings were observed: i) the selection of the optimal parameter value strongly depends on the specific experimental conditions considered, ii) in 15 out of 18 conditions the errors obtained using the proposed approach and the trial-and-error were coincident, while in the other cases the maximum discrepancy amounted to 2.5 deg and less than 1.5 deg on average.

Orientation Estimation Through Magneto-Inertial Sensor Fusion: A Heuristic Approach for Suboptimal Parameters Tuning / Caruso, Marco; Sabatini, Angelo Maria; Knaflitz, Marco; Gazzoni, Marco; Croce, Ugo Della; Cereatti, Andrea. - In: IEEE SENSORS JOURNAL. - ISSN 1530-437X. - 21:3(2021), pp. 3408-3419. [10.1109/JSEN.2020.3024806]

Orientation Estimation Through Magneto-Inertial Sensor Fusion: A Heuristic Approach for Suboptimal Parameters Tuning

Caruso, Marco;Knaflitz, Marco;Gazzoni, Marco;Cereatti, Andrea
2021

Abstract

Magneto-Inertial Measurement Units (MIMUs) are a valid alternative tool to optical stereophotogrammetry in human motion analysis. The orientation of a MIMU may be estimated by using sensor fusion algorithms. Such algorithms require input parameters that are usually set using a trial-and-error (or grid-search ) approach to find the optimal values. However, using trial-and-error requires a known reference orientation, a circumstance rarely occurring in real-life applications. In this article, we present a way to suboptimally set input parameters, by exploiting the assumption that two MIMUs rigidly connected are expected to show no orientation difference during motion. This approach was validated by applying it to the popular complementary filter by Madgwick et al. and tested on 18 experimental conditions including three commercial products, three angular rates, and two dimensionality motion conditions. Two main findings were observed: i) the selection of the optimal parameter value strongly depends on the specific experimental conditions considered, ii) in 15 out of 18 conditions the errors obtained using the proposed approach and the trial-and-error were coincident, while in the other cases the maximum discrepancy amounted to 2.5 deg and less than 1.5 deg on average.
File in questo prodotto:
File Dimensione Formato  
Caruso et al_2021.pdf

non disponibili

Tipologia: 2a Post-print versione editoriale / Version of Record
Licenza: Non Pubblico - Accesso privato/ristretto
Dimensione 3.02 MB
Formato Adobe PDF
3.02 MB Adobe PDF   Visualizza/Apri   Richiedi una copia
Caruso 2020_IEEE_sensors_postprint.pdf

accesso aperto

Tipologia: 2. Post-print / Author's Accepted Manuscript
Licenza: PUBBLICO - Tutti i diritti riservati
Dimensione 869.73 kB
Formato Adobe PDF
869.73 kB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2861872