Description du sujet de thèse

This thesis aims to develop a subject-specific model predictive control approach for assistive robots in the
context of teaching/rehabilitating upper limb motions to visually and physically impaired patients. The
assistive robot will be composed of two torque-controllable serial manipulators, allowing an unprecedented
flexibility in the assistive motion prescriptions. Recent progress in control, stemming from highly redundant
humanoid robots developed at LAAS-CNRS, as well as inverse Reinforcement Learning (IRL) and motion
analysis, developed at IPAL and NUS, will be adapted to improve patient/therapist-specific human-robot
interaction.

Contexte de travail

The PhD thesis will be co-supervised by researchers at LAAS-CNRS (France), IPAL (Singapore), and NUS
(Singapore). In France, Senior Research Director Florent Lamiraux will be the official PhD supervisor; he is
an expert in planning and humanoid robotics. In Singapore, Prof. Harold Soh (NUS) will be a co-supervisor;
he is a world-renowned expert in robot learning and planning. Vincent Bonnet (IPAL) will also be a co-
supervisor, with expertise in collaborative robotics and human motion analysis and prediction. The Director
of IPAL, Christophe Jouffrais, is a world leader in developing aids for visually impaired patients. The PhD
student is expected to spend at least 1 year in Singapore.

Le poste se situe dans un secteur relevant de la protection du potentiel scientifique et technique (PPST), et nécessite donc, conformément à la réglementation, que votre arrivée soit autorisée par l'autorité compétente du MESR.