Missions

The missions of this postdoc is to develop a force enabled operator-centred Model Predictive Controller of a collaborative robot to reduce their musculoskeletal disorders.

Activités

The person is expected to develop a new theoretical framework for force controlled collaborative robot working in physical interaction with a human. The work will consist in reading the literature and improve the state of the art already developed by the team. Then the proposed method will have to be implemented in C++/Python and benchmark with other state of the art methods such as Reinforcement Learning and Model Predictive Path Integral. Once validated in simulation the proposed controller will have to be tested on a Panda collaborative robot with human collaborator. To do so a “Comité de Protection des Personnes” will have to be written and submitted at the beginning of the postdoc.

Compétences

Proficiency in programming in C++, Python,
Basic skills in machine learning and PyTorch, experience with GPU is a plus
Expertise in optimal control and non linear optimization
Hands on expertise in ROS control
Ability to work and write scientific paper in English

Contexte de travail

Work in the Gepetto team of LAAS-CNRS, in the context of the ANR project HERCULES, with LIRMM

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.

Contraintes et risques

No risk