> Open positions

The lab is located in Nancy, France, a nice and lively city in the East of France. The fast train (TGV) allows us to reach the center of Paris in 1.5 hours (Garde de l'Est). The cost of life is much lower in Nancy than in Paris (or other similar cities, like London).

Post-doc: combining QP-based whole body control and trial-and-error learning

We are is seeking for a highly qualified postdoctoral researcher with strong interests in one or more of the following research topics:
  • machine learning for robotics (especially, direct policy search in reinforcement learning);
  • legged locomotion;
  • optimal control theory

Please relate clearly to at least one or more of these topics in your Research Statement. The candidates are expected to contribute to the ResiBots ERC project (see below). Successful candidates can furthermore be given the opportunity to work with undergraduate, M.Sc. and Ph.D. students.

The main robots will be our 6-legged robots and a new, highly-dynamic quadruped robot.

Ph.d. students

If you are a potential Ph.D. student with the possibility to obtain your own funding, send me a research proposal and a CV.

Master/engineer internships

We hire a few interns every year to work on our research projects. Please send a resume and a cover letter that describes why you are interested in our research topic and your background (in French or in English).

All internships are paid about 500 euros / month. They are full-time positions in the lab. We can help you to find an accomodation in Nancy. The official language of the team is English: English-speaking international students are welcomed.

Below are some topics for 2017/2018. Please refer to them if you apply.

[mechatronics] Design of centimeter-scale robots for archeology

We are collaborating with the HIP institute to develop novel robotic approaches for archaeology. The objective of this internship is to help us to design a new centimeter scale robot for minimally invasive exploration. The intern will work on some sub-parts of the system to be determined, depending on the current state of the project.

About the project: Responsibilities:
  • search and develop innovative mechanical solutions for different requirements of the robot;
  • evaluate the performance, by making prototypes combining different manufacturing methods, like high resolution 3D printing and conventional manufacturing (CNC milling).
Useful skills:
  • Good knowledge of mechanical design with Solid Works (Design, Drawing, Simulation, FEM)
  • 3D printing and conventional manufacturing
  • General electronic knowledge
  • Arduino programming (optional)
Typical profile:
  • Student in Engineer school or DUT specialized in mechanics
  • Proactive
  • Meticulous and exigent (soldering, assembling small parts)
  • Good communication skills in English (the official language)


[mechatronics] Improving a wheel-leg hybrid robot for versatile adaptation

Our team works on new machine learning algorithms to allow robots to adapt to unexpected situations. For instance, we seek to make a hexapod robot that broke a leg re-learn to walk in a few minutes. We recently obtained promising results (see [1] and [2]).

We now wish to extend the adaptation possibilities by making our approach work on an hybrid wheel-legs robot, that is to say a robot capable of both walking and driving (see [3]). To do so, a new gait controller has to be designed and our learning algorithms have to be adapted.

The objective of this internship is:
  • improve the existing wheel-legs platform (hardware and software);
  • adapt the walking and learning algorithms to use the wheels;
  • extend the current hexapod simulator to handle wheels.

The position is for at least 5 months and, due to the French law, available only to students.

We aim at experimenting on the robots as much as possible. In case of blocking hardware issue, the work could be done in simulation.

References:
  • [1] http://www.resibots.eu/videos.html
  • [2] Cully, A., Clune, J., Tarapore, D., & Mouret, J. B. (2015). Robots that can adapt like animals. Nature, 521(7553), 503-507.
  • [3] http://www.resibots.eu/videos.html#the-creadapt-robot-6-legged-hybrid

Depending on the selected candidate, the internship can be more focused on the mechatronics side or the machine learning side. Nonetheless, basic skills are wished in:

  • C++ programming,
  • Python programming,
  • use of ROS,
  • basic CAD (we use SolidWorks or OpenSCAD),
  • English
  • (optional) machine learning.