Data Fusion-Based Musculoskeletal Synergies in the Grasping Hand

Abstract

The hypothesis that the central nervous system (CNS) makes use of synergies or movement primitives in achieving simple to complex movements has inspired the investigation of different types of synergies. Kinematic and muscle synergies have been extensively studied in the literature, but only a few studies have compared and combined both types of synergies during the control and coordination of the human hand. In this paper, synergies were extracted first independently (called kinematic and muscle synergies) and then combined through data fusion (called musculoskeletal synergies) from 26 activities of daily living in 22 individuals using principal component analysis (PCA) and independent component analysis (ICA). By a weighted linear combination of musculoskeletal synergies, the recorded kinematics and the recorded muscle activities were reconstructed. The performances of musculoskeletal synergies in reconstructing the movements were compared to the synergies reported previously in the literature by us and others. The results indicate that the musculoskeletal synergies performed better than the synergies extracted without fusion. We attribute this improvement in performance to the musculoskeletal synergies that were generated on the basis of the cross-information between muscle and kinematic activities. Moreover, the synergies extracted using ICA performed better than the synergies extracted using PCA. These musculoskeletal synergies can possibly improve the capabilities of the current methodologies used to control high dimensional prosthetics and exoskeletons.

Biomimetic learning of hand gestures in a humanoid robot

“…Leveraging the insights from our previous works ( Olikkal et al, 2022a , b ) it was observed that time-variant synergy models yielded best results, hence for this study, we opted for the time-variant synergy model. In case of time-variant models, a time-varying pattern is produced by combining synergies with time-varying scaling coefficients.…”

“…Principal component analysis (PCA) was applied on the cascaded velocity matrix to derive PCs that capture the maximum variance. Following our prior works ( Olikkal et al, 2022a , b ), to identify the optimal number of PCs, we selected those PCs the accounted for 0.98 (98%) of total variance using the Eq. (3) expressed as…”

“…Our goal is to replicate this motor learning in a robot, providing insights into the neural mechanisms of motor planning and execution along with their limitations. Building on the evidence from previous studies ( Vinjamuri et al, 2010 ; Patel et al, 2018 ; Pei et al, 2019 ; Olikkal et al, 2022a ), which leverage electroencephalography (EEG), electromyography (EMG), and hand joint kinematics to identify neural mechanisms of motor planning and motor execution, we aim to understand and replicate these synergies in a robotic system. Burns et al (2017) contributed to this understanding by developing a soft hand exoskeleton – a form of human-machine teaming – embedding motor synergies for assisting or rehabilitating individuals with hand disabilities.…”

“…Research study by Jarque-Bou et al (2019b) adopted a phase-based approach, decomposing the movement profile into distinct stages: reaching, grasping/manipulation and release ( Figure 1 ) in an attempt to understand how the CNS controls and coordinates the hand when aiming to execute a targeted movement. The joint angular velocity profiles of these three phases collected from joint recording devices such as CyberGlove are well represented and investigated in Vinjamuri et al (2010) , Jarque-Bou et al (2019a , b) , and Olikkal et al (2022a , b) . These studies reinforce that when a subject attempts to reach a target from initial reference position, a progressive increase in angular velocity profile can be observed in the reaching phase.…”