Assistive robotics: Semi-autonomous definition of ARPH mobile platform configuration

Issue title: Virtual Reality and Disability

Guest editors: Christian Berger-Vachonx, Daniel R. Mestre and Alain Pruskiy

Article type: Research Article

Authors: Rezzoug, Nasser^a | Gorce, Philippe^a | Hoppenot, Philippe^b | Colle, Etienne^b

Affiliations: [a] Laboratoire ESP EA 3162, Université du Sud Toulon, Var, France | [b] Laboratoire Systèmes Complexes, Université d'Évry Val d'Essonne, France | [x] University of Lyon, France | [y] Université François-Rabelais de Tours, Laboratoire d'Informatique, Polytech Tours, France

Correspondence: [*] Address for correspondence: Philippe Gorce, Avenue de l'université, BP 20132, 83957 La Garde cedex, France. Tel.: +33 4 94 14 27 55; Fax: +33 4 94 14 22 78; E-mail: [email protected].

Abstract: In this paper, we present a learning based method for the definition of the ARPH mobile platform configuration during grasping. Based on the principle of semi autonomy, the proposed scheme uses information provided by the user to help him(her) during the tedious task of manually coordinating the robot motions. From a crude posture provided by the user a reinforcement scheme is used to optimize the kinematic configuration of the MANUS robot to align the gripper with the axis of the object and grasp it. Simulation results illustrate the capability of the proposed method.

Keywords: Mobile platform, navigation, grasping, semi-autonomy, neural network

DOI: 10.3233/TAD-2006-18405

Journal: Technology and Disability, vol. 18, no. 4, pp. 189-194, 2006