Simulation-based Shaping of Deformable Objects in Collaborative Robotic Applications

We are recruiting a student at LCFC (Metz, France) for a PhD in the ANR Mandolin project on the topic "Simulation-based Shaping of Deformable Objects in Collaborative Robotic Applications". More information is given below and in the attached proposal.

-- Scientific context --
Deformable objects are ubiquitous in industrial and household applications. However, it is still difficult for robot manipulators to handle them properly. The robots are generally not aware of the deformation physics at play, so they cannot control the shape or ensure the object’s integrity at all times. This limitation gave rise to the field of deformable object manipulation, which seeks to improve the robot’s ability to manipulate soft objects. Some promising solutions consist in leveraging fast mechanical simulations as an internal model for a motion controller. This strategy was also adopted in robotized medical surgery for instance.

-- Challenges --
• Formulate control algorithms that take advantage of the interactive simulation, to improve the controller’s performance;
• Perform online deformation state estimation and controller adaptation, to mitigate the mismatch between simulated and perceived quantities;
• Define collaboration modalities in the shaping task, to investigate multi-robot and human-robot applications.

The PhD is a part of the activities of the Control team of the LCFC, where we are interested in robotic systems design and robotization of manufacturing processes. You will integrate a human-scaled work environement, boasting ambitious projects and state-of-the-art experimental platforms.

Candidates should have a Master’s degree (M2 or equivalent) with research experience in the field of robotics, mechatronics, or mechanical engineering. We seek the follwing skills:
• Languages: written and spoken English, scientific writing. Spoken French is a plus.
• Robotics: kinematics, motion planning, motion control.
• Control: linear systems, Lyapunov synthesis, adaptive control.
• Mechanics: solid dynamics, continuum mechanics, finite element modeling.
• Tools: programming in C++ and/or Python, knowledge in ROS2.
• Knowledge in computer vision or artificial intelligence is a plus.
• Soft skills: will to learn, autonomy, palatability for project-oriented organization.