A formal methodology for avoiding hyperstaticity when connecting an exoskeleton to a human member - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2010

A formal methodology for avoiding hyperstaticity when connecting an exoskeleton to a human member

Résumé

The design of a robotic exoskeleton often focuses on replicating the kinematics of the human limb that it is connected to. However, human joint kinematics is so complex that in practice, the kinematics of artificial exoskeletons fails to reproduce it exactly. This discrepancy results in hyperstaticity. Namely, uncontrolled interaction forces appear. In this paper, we investigate the problem of connecting an exoskeleton to a human member while avoiding hyperstaticity; to do so, we propose to add passive mechanisms at each connection point. We thus introduces a formal methodology for avoiding hyperstaticity when connecting wearable robotic structures to the human body. First, analyzing the twist spaces generated by these fixation passive mechanisms, we provide necessary and sufficient conditions for a given global isostaticity condition to be respected. Then, we derive conditions on the number of Degrees of Freedom (DoFs) to be freed at the different fixations, under full kinematic rank assumption. We finally apply the general methodology to the particular case of a 4 DoF shoulder-elbow exoskeleton. Experimental results allow to show an improvement in transparency brought by the passive mechanism fixations.
Fichier principal
Vignette du fichier
2010ACTI1514.pdf (657.1 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02110635 , version 1 (25-04-2019)

Identifiants

  • HAL Id : hal-02110635 , version 1

Citer

Nathanael Jarrasse, Guillaume Morel. A formal methodology for avoiding hyperstaticity when connecting an exoskeleton to a human member. 2010 IEEE International Conference on Robotics and Automation (ICRA 2010), May 2010, Anchorage, France. pp.1188-1195. ⟨hal-02110635⟩
32 Consultations
75 Téléchargements

Partager

Gmail Facebook X LinkedIn More