Post doctorat

Postdoc position: Control of a robot manipulator to assist people with disabilities

Équipes / Services concernés

Responsables

Martin Mujica

Date de publication

13.01.25

Prise de poste souhaitée

15.02.25

Practical information:

Contract: 24 months (CDD)

Starting date: Immediate or based on candidate availability

Salary: Between 2991€ and 3417€ depending on experience

Location: LAAS-CNRS, Toulouse, France

Studies: PhD in the domains of robotics and control.

Context:

Recent years have shown the development of collaborative robots, where the robots’ design has been adapted, as well as the approaches for robot control, to interact with people around them. The EXTENDER project aims at evaluating and adapting recent control techniques for collaborative robot manipulators, to apply them for assistive technology for people with disabilities.

The project specifically focuses on leveraging state-of-the-art control techniques, developed and tested in laboratory settings, to assist individuals with limited arm or hand mobility in performing complex tasks. In fact, collaborative robots offer a unique opportunity to enhance autonomy and improve quality of life, provided they are designed with a user-centered approach. Therefore, the control laws will be adapted and applied to a robot mounted on a wheelchair, to react to the users input, allowing the person to perform real life tasks.

This project is part of a technological transfer initiative from research laboratories to industrial partners. As such, the outcomes of the project are not limited to academic contributions; they hold the potential for real-world applications that can directly improve the lives of people with disabilities or limited mobility.

Objective and missions:

The recruited postdoc will be involved in adapting state of the art control methods in order to be used on the new robot arm from Orthopus (a project’s partner), that will be mounted on a wheelchair. As the inputs from the person might come in different modalities (depending on the person’s disability) the control of the robot should be able to adapt to produce a safe and appropriate output from the manipulator. Furthermore, it is expected that the user on the wheelchair can control the robot to perform challenging tasks, while preserving the safety of the person.

The chose candidate will study the state of the art of current solutions for the proposed case, with the particularities related to assistance robotics. Then, the development of control techniques based on adaptable and robust admittance control [1] to allow the robot interact with the environment based on the person’s inputs is expected. Furthermore, as the environment will also evolve, Optimal control techniques based on vision and forces [2] are to be evaluated in this context too, using real time libraries [3].

The postdoctoral researcher will work on these subjects to implement them (with the help of the teams’ engineers) on our existing robot manipulators (Franka Emika) as well as to the newly developed robot from Orthopus. The results obtained of these advanced control laws, while applied on a new manipulator for real life complex tasks, are expected to be published. It is also expected that the recruited person will discuss, collaborate and interact with the other partners of the project in order obtain reliable and reusable solutions.

Working environment:

This position is part of France 2030 project within a consortium of laboratories as well as medical and industrial partners. The recruit will work within the RAP team (https://www.laas.fr/en/teams/rap/) at LAAS-CNRS in collaboration with a permanent team member and an engineer from the team. The RAP team is composed of 17 people between permanents, postdoc, engineers and PhD students. The expertise of the team goes from multi-sensor perception to sensor-based control of robots in dynamical environments. This makes a good environment for the recruited person to interact and collaborate with the other members. As mentioned, the recruited person will work for the EXTENDER project, which will involve a close interaction with people from the other partners in France (meetings, code-sharing, demos).

The team is part of a leading robotics department with several robotic platforms and is engaged in multiple cutting-edge projects. The laboratory itself has approximately 600 people, providing an interesting and dynamic environment for personal and professional development.

Expected profile:

This position is intended for candidates holding a PhD degree in the field of robotics and control, with a strong knowledge of standard programming languages. The expected skills from the candidate are:

  • Modeling and control of robots
  • C++ and Python programming experience
  • Use of ROS as middleware for robotics and Git for collaborative projects
  • Writing proficiency in English
  • Autonomy and teamwork
  • Scientific curiosity

How to apply:

In order to apply, the person should send a single PDF file to containing:

  • Candidate’s CV
  • Motivation letter
  • 1 or 2 references with contact information and if possible, a recommendation letter

[1] Mujica, M., Crespo, M., Benoussaad, M., Junco, S., & Fourquet, J. Y. (2023). Robust variable admittance control for human–robot co-manipulation of objects with unknown load. Robotics and Computer-Integrated Manufacturing, 79, 102408.

[2] Gebrayel, F., Mujica, M., & Danès, P. (2024, November). Visual Servoing for Vine Pruning based on Point Cloud Alignment. In Icinco.

[3] Mastalli, C., Budhiraja, R., Merkt, W., Saurel, G., Hammoud, B., Naveau, M., ... & Mansard, N. (2020, May). Crocoddyl: An efficient and versatile framework for multi-contact optimal control. In 2020 IEEE International Conference on Robotics and Automation (ICRA)