PhD Position F/M Campagne/Welcome Package - Learning robotic manipulation at scale

The work will be conducted in the WILLOW team at Inria Paris research center. Renowned for its exceptional work in computer vision and robotics, the WILLOW team has consistently produced high-quality research, resulting in publications in major journals and conferences.

As part of the team, you will have access to a well-established laboratory featuring multiple robotic arms, hands, quadrupeds, bipeds, and mobile manipulators.

Additionally, you can expect frequent visits and talks by esteemed researchers from top research laboratories around the world. Opportunities abound for collaboration with leading researchers both in Europe and globally.

Furthermore, you will join an international and welcoming team environment, where we regularly organize various events ranging from casual after-work gatherings to multi-day lab retreats. 

Assignments :

Imagine a household robot capable of performing various chores, from cleaning houses to folding clothes, rearranging items and even cooking meals. While existing robotic hardware is able to handle such diverse tasks, e.g., through tele-operation, a critical gap exists in the development of models that enable robots to autonomously execute these activities and generalize across different environments. The goal of the position is to bridge this gap by learning robotic policies to perform a wide range of daily tasks in real-world environments. 

Data plays an important role for robot learning. The data typically used in training can be divided into three categories: real robot data, simulated robot data and real-world Internet data. The real robot data requires many robots and human tele-operation to collect, remaining limited in scale and diversity to learn various robot skills. The simulated robot data is easier to construct, but still requires much engineering work to create scripts and suffers from a sim-to-real gap. The real-world Internet data though is large-scale and contains real-world interactions, it does not have action annotations for robot training.

In this project, the candidate will explore different ways to overcome the data limitations to scale up robot skill learning, including but not limited to: 1) utilizing robot play data to avoid cost of human tele-operation; 2) taking advantage of foundation models to automatically generate simulated data; 3) learning policies from human-object interaction videos.

Supervisors :

Shizhe Chen, Cordelia Schmid

• Read papers
• Propose methods
• Conduct experiments
• Analyze results
• Write papers
• Present work in conferences
• Write a doctoral thesis

The candidate must have an excellent track of records and a Master Degree (or equivalent). The candidate must have the following qualifications:

• Strong background in computer vision, robotics, or related fields
• Excellent programming skill in Python and Pytorch
• Strong proficiency in both written and spoken English
• Ability to work independently as well as collaboratively
• Ability and desire to read prior work and to build upon it in one’s own work.

• Subsidized meals
• Partial reimbursement of public transport costs
• Leave: 7 weeks of annual leave + 10 extra days off due to RTT (statutory reduction in working hours) + possibility of exceptional leave (sick children, moving home, etc.)
• Possibility of teleworking
• Flexible organization of working hours (after 12 months of employment) 
• Professional equipment available (videoconferencing, loan of computer equipment, etc.)
• Social, cultural and sports events and activities
• Access to vocational training
• Social security coverage

According to civil service salary scales