Design, implementation and control of a deformable manipulator robot based on a compliant spine - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue The International Journal of Robotics Research Année : 2020

Design, implementation and control of a deformable manipulator robot based on a compliant spine

Résumé

This paper presents the conception, the numerical modeling and the control of a dexterous, deformable manipulator bio-inspired by the skeletal spine found in vertebrate animals. Through the implementation of this new manipulator, we show a methodology based on numerical models and simulations, that goes from design to control of continuum and soft robots. The manipulator is modeled using Finite Element Method (FEM), using a set of beam elements that reproduce the lattice structure of the robot. The model is computed and inverted in real-time using optimisation methods. A closed-loop control strategy is implemented to account for the disparities between the model and the robot. This control strategy allows for accurate positioning, not only of the tip of the manipulator, but also the positioning of selected middle points along its backbone. In a scenario where the robot is piloted by a human operator, the command of the robot is enhanced by a haptic loop that renders the boundaries of its task space as well as the contact with its environment. The experimental validation of the model and control strategies is also presented in the form of an inspection task use case.
Fichier principal
Vignette du fichier
IJRR_EchelonIII_corrected (2).pdf (23.02 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03028723 , version 1 (11-01-2021)

Identifiants

Citer

Thor Morales Bieze, Alexandre Kruszewski, Bruno Carrez, Christian Duriez. Design, implementation and control of a deformable manipulator robot based on a compliant spine. The International Journal of Robotics Research, 2020, ⟨10.1177/ToBeAssigned⟩. ⟨hal-03028723⟩
150 Consultations
212 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More