PhD Position F/M Learning Sequences of Contact States for Object Manipulation – Montbonnot-Saint-Martin, 38330

PhD Position F/M Learning Sequences of Contact States for Object Manipulation

Inria

Montbonnot-Saint-Martin, 38330

Postuler

PhD Position F/M Learning Sequences of Contact States for Object Manipulation

Le descriptif de l’offre ci-dessous est en Anglais

Type de contrat : CDD

Niveau de diplôme exigé : Thèse ou équivalent

Fonction : Doctorant

A propos du centre ou de la direction fonctionnelle

The Inria Grenoble research center groups together almost 600 people in 23 research teams and 7 research support departments.

Staff is present on three campuses in Grenoble, in close collaboration with other research and higher education institutions (University Grenoble Alpes, CNRS, CEA, INRAE, …), but also with key economic players in the area.

Inria Grenoble is active in the fields of high-performance computing, verification and embedded systems, modeling of the environment at multiple levels, and data science and artificial intelligence. The center is a top-level scientific institute with an extensive network of international collaborations in Europe and the rest of the world.

Contexte et atouts du poste

Collaboration between Inria team THOTH and the multidisciplinary institute in artificial intelligence (MIAI) in Grenoble

Research activities in MIAI aim to cover all aspects of AI and applications of AI with a current focus on embedded and hardware architectures for AI, learning and reasoning, perception and interaction, AI & society, AI for health, AI for environment & energy, and AI for industry 4.0.

This project in particular focuses on perception and robot interaction and thus will take place in close collaboration with LIG and GIPSA-Lab at the University of Grenoble.

Mission confiée

Key words

motion planning, trajectory prediction, manipulation, contact

Theme/Domain

Learning Sequences of Contact States for Object Manipulation 

Context and motivation

Humans interact with objects and their environment with seemingly very low effort. A large portion of decisions about the motion path, contact and the 3D scene arises from past experience about interactions. Evidence from the cognitive science research state that in particular making contact plays a central role in understanding human-object interactions [1, 2]. In this project we are interested in learning robot-object interactions and how those emerge from a sequence of contact states with the environment.

Contact

[email protected]
Please apply with the typical application documents: CV, certificates (MSc degree), research statement and two references. Application deadline is November 30th. We will screen applications on a rolling basis. The position will be filled as soon as a suitable candidate is found.

Main location

INRIA Montbonnot, TOTH team

Principales activités

Summary

In robotics many researchers have explored individual skills like pushing, pivoting, manipulation, grasping at depth. This project aims at estimating opportunities of contact that automatically synthesise such motion primitives. 

Approach

Efficient manipulation requires contact to reduce uncertainty but maybe even more important to synthesise effective motion trajectories with simple grippers. Taking the external environment into account for the manipulation process (“extrinsic dexterity”) allows to achieve complex manipulation behaviour also with simple grippers that only have two or even just one point grippers. The key idea of this project is to make use of visually detected contact modes of environment-object contacts to guide the evolution of the motion trajectories. To this end this project comprises two main parts – the visual detection of contact opportunities offered by the environment and the generation of optimised contact interactions that are guided by detected contact states.

GOAL

Developing novel representations of contact for robot manipulation tasks. Modelling contact opportunities with the environment will allow fundamentally new ways of interpreting grasps as dynamic motion trajectories that are not only guided by pre-computed static hand-object configurations, further more those allow for incorporating essential information about the path leading to a physically plausible and (temporally) stable hand-object-environment configuration. 

References

[1] Zago, M., McIntyre, J., Senot, P., & Lacquaniti, F. (2009). Visuo-motor coordination and internal models for object interception. Experimental Brain Research, 192, 571-604. 

[2] Tresilian, J. R. (1995). Perceptual and cognitive processes in time-to-contact estimation: Analysis of prediction-motion and relative judgment tasks. Perception & Psychophysics, 57(2), 231-245.

[3] Toussaint, M., Ratliff, N., Bohg, J., Righetti, L., Englert, P., & Schaal, S. (2014, September). Dual execution of optimized contact interaction trajectories. In 2014 IEEE/RSJ International Conference on Intelligent Robots and Systems (pp. 47-54). IEEE.

[4] Cheng, X., Huang, E., Hou, Y., & Mason, M. T. (2021, May). Contact mode guided sampling-based planning for quasistatic dexterous manipulation in 2d. In 2021 IEEE International Conference on Robotics and Automation (ICRA) (pp. 6520-6526). IEEE.

[6] Halawa, M., Hellwich, O., & Bideau, P. (2022, October). Action-based contrastive learning for trajectory prediction. In European Conference on Computer Vision (pp. 143-159). Cham: Springer Nature Switzerland.

[7] Zhou, W., & Held, D. (2023, March). Learning to grasp the ungraspable with emergent extrinsic dexterity. In Conference on Robot Learning (pp. 150-160). PMLR.

Compétences

  • MSc degree in computer science or similar field
  • Experience in robotics, machine learning, computer vision, and/or control, industrial experience is a plus.
  • Experience in applying (deep) learning to robotic control problems
  • Excellent software engineering and programming skills in C++ and/or Python
  • Interest in interdisciplinary research in the context of the MIAI Grenoble Alpes Institute
  • Excellent English writing and communication skills

Avantages

  • 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 (90 days / year) and flexible organization of working hours
  • Professional equipment available (videoconferencing, loan of computer equipment, etc.)
  • Social, cultural and sports events and activities
  • Access to vocational training
  • Social security coverage under conditions

Rémunération

1st and 2nd year: 2 082 euros gross salary /month



3rd year: 2 190 euros gross salary / month

Postuler

Voir tous les emplois