Home » Join us » Opportunities

Opportunities

Job offers

Position: University Professor “Artificial Intelligence: Theory and Applications

The position is open to all areas of AI and its applications. The successful candidate will integrate one of the laboratories: ISIR, LIMICS or LIP6 according to his/her research themes, and/or projects involving several host laboratories within SCAI (Sorbonne Center for Artificial Intelligence). The professor should be able to coordinate national and international collaborative programmes. The candidate’s past participation in multidisciplinary projects will be appreciated.

The person recruited will contribute significantly to the teaching of the Bachelor of Computer Science whose needs cover the whole discipline (algorithms, programming in particular object, concurrent, functional, web), discrete mathematics, data structures, systems, architecture, networks, compilation, databases…) as well as to the Master of Computer Science, in particular for the ANDROIDE, BIM or DAC courses.

More details and the complete job description to come on :
https://recrutement.sorbonne-universite.fr/fr/personnels-enseignants-chercheurs-enseignants-chercheurs/enseignants-chercheurs/recrutement-2022-des-enseignantes-chercheuses-et-enseignants-chercheurs.html

Contact at ISIR: Guillaume Morel ; guillaume.morel(at)sorbonne-universite.fr

Download the offer

Profile:

The desired profile is that of a roboticist with a pronounced taste for experimental activities. The candidate must demonstrate a mastery of dynamic analysis and the control of manipulative and mobile robots. Skills in the entire chain of implementation of control laws, such as data, signal and image processing will also be appreciated.

The willingness to participate in educational innovation projects would be a plus.

In view of the subjects to be taught, a mastery of computer tools allowing the implementation of control laws and perception (ROS, C ++, Python) is required.

Training involved:

The new recruit will be involved in the EEA Licence, in the Automation & Robotics Master’s degree as well as in the various specialities of Polytech Sorbonne.

According to her/his profile, she/he will participate in the robotics initiation and automation programming courses in the complementary trades minor of the EEA Bachelor’s degree, as well as in electronics in L1. She will reinforce the teams in charge of teaching mobile robotics, robotic manipulators, automation and signal processing in the master’s degree and at Polytech Sorbonne.

Research :

Profile, team or research unit planned, or emerging discipline or innovation in line with the research component of the institution’s four-year contract.

The successful candidate will conduct research at ISIR in one of the fields related to robotic manipulation: dexterous manipulation, collaborative manipulation, multi-scale manipulation, telemanipulation, comanipulation, exoskeletons and upper limb prostheses, etc. This research is intended to be integrated as a priority into the SYROCO team. However, this priority is not absolute and any application proposing research in manipulative robotics with integration in another team will be examined with attention.

The research will be conducted in system and interaction modelling, the exploitation of these models for the synthesis of commands or behaviours, and the processing of data provided by sensors for the real-time implementation of these command or behaviour laws. The ability to validate the work on demonstrators is an important asset for the position.

Research contact:

Guillaume MOREL, Director of ISIR: guillaume.morel(at)sorbonne-universite.fr

Download the offer

Postdoctoral or research engineer positions for the HCI Sorbonne group (Human Computer Interaction)

Context:

We have multiple postdoctoral and engineering positions in the HCI group (https://hci.isir.upmc.fr) at Sorbonne Université, Paris, France.

Missions:

We are searching for curious minds who want to conduct cutting-edge research at the intersection of HCI with VR, Haptics, Robotics or AI. Possible topics/areas of research are:

  • Novel Interaction techniques in VR,
  • VR and haptics for gaming or training,
  • Computational models of decision-making and human performance,
  • AI-based recommendation systems,
  • Tele-operation and remote collaboration.

Some of our previous work in these areas:

Required profile:

For the postdoctoral position, a Phd degree in Computer science, HCI or other field related to our research areas is required.

Required skills:

  • strong programming and analytical skills,
  • strong background in at least one of the following areas (HCI, VR, Haptics, Robotics, AI).

More information : 

  • Type of position: Postdoctoral or Research Engineer position
  • Start date: as soon as possible
  • Duration: 1 to 2 years
  • Level of study required: Master 2 (for engineer), PhD (for post-doc)
  • Location: ISIR (Institut des Systèmes Intelligents et de Robotique), Campus Pierre et Marie Curie, 4 place Jussieu, 75005 Paris.

Contacts: 

  • Gilles Bailly et Sinan Haliyo
  • Email : gilles.bailly(at)sorbonne-universite.fr ; sinan.haliyo(at)sorbonne-universite.fr
  • Application: Send your application by email, with a CV and a cover letter.
  • Application deadline: None

Download the offer

Internship offers

Subject: Design and control of an articulated rover

Abstract:

Within the framework of planetary or agricultural applications, an autonomous robot is faced to difficulties of overcoming obstacles (positive or negative), irregular soils and/or slopes and slopes. The aim is to design an agile robot with wheels and legs and with advanced locomotion capabilities.

Internship Objectives:

The first part of the internship consists in designing the virtual model in SolidWorks. The kinematic concept and the general architecture are more or less decided, as well as the choice of the electric motors.

The second part concerns the dynamic modeling and the control. The modeling must be able to describe the dynamics of the center of gravity and its angular momentum, known as centroidal dynamics. This model makes it possible to define a control in impedance to ensure the stability of the rover and the follow-up of trajectories and to deduce finally the joint torques in the driving joints. These commands must be tested on the virtual model imported under ROS-Gazebo.

During this period of modeling-command, we will launch the manufacturing of the parts. We will start by validating the concept of a wheel leg before proceeding to the assembly of the whole.

Required Profile: Master level 2, on Mechatronics-Robotics

Required skills: Design, modeling and control, robotics fundamentals

 

  • Supervisor: Faïz BEN AMAR
  • Start date: 1st February ou 1st March
  • Duration : 5/6 months
  • Level of study required: Master 2 or final year of engineering school
  • Location: ISIR (Institut des Systèmes Intelligents et de Robotique), Campus Pierre et Marie Curie, 4 place Jussieu, 75005 Paris.

Contact person:

  • Faïz BEN AMAR
  • Tel : +33 1 44 27 63 42
  • Email : amar(at)isir.upmc.fr
  • Send your application by email, with [internship subject] in the subject line, with CV and motivation letter.
  • Application deadline: 28 February 2022

Download the internship offer

Subject: Meta-learning for divergent exploration with sparse rewards

Abstract:

Quality-Diversity (QD) methods [2, 3], whose aim is to discover behaviorally diverse sets of solutions to problems, have proven to be effective tools in dealing with sparse rewards in Reinforcement Learning. However, they struggle with generalization to unseen tasks, for which they often need to be trained from scratch. The goal of this internship will be to investigate the use of prior learning experience to improve the learning process itself, in a manner that is closely related to optimization-based meta-learning approaches such as [4].

This internship addresses problems that are relevant to many industrial applications such as robotics manipulation and open-ended learning. In particular, it will be part of the European project Veridream [1] which involves multiple industrial partners.

Internship Objectives:

The questions that will be investigated during this internship are in direct continuity with [6, 5]. The successful candidate will

  • be responsible for the evaluation of existing approaches, in particular with regard to the dynamics of populations of agents,
  • contribute to the design and evaluation of methods that quantify and address variability in the optimization process.

Depending on the progress made on those subjects as well as the candidate’s interests, other research directions in relation to robustness and generalization will be considered.

Required Profile:

The candidate should have a strong interest in Machine Learning and be enrolled in a MSc or engineering school program in Computer Science, Machine Learning or related fields.

Required skills :

A strong mathematical background, good development skills and proficiency in the Python programming language are mandatory. A working knowledge of English is required, knowledge of French is appreciated but not necessary.

 

  • Supervisors: Achkan Salehi, Stéphane Doncieux et Alex Coninx
  • Duration: 6 months
  • Level of study required: Master 2 or engineering school 3rd year
  • Location: ISIR (Institut des Systèmes Intelligents et de Robotique), Campus Pierre et Marie Curie, 4 place Jussieu, 75005 Paris.

 

Contact person:

  • Achkan Salehi
  • Tel: +33 1 44 27 63 82
  • Email: salehi(at)isir.upmc.fr
  • Send your application by email, with [internship subject] in the subject line, with CV and motivation letter

Download the intership offer

 

References :
[1] https://www.veridream.eu/.
[2] A. Cully and Y. Demiris. Quality and diversity optimization: A unifying modular framework. IEEE Transactions on Evolutionary Computation, 22(2):245–259, 2017.
[3] S. Doncieux, A. Laflaquière, and A. Coninx. Novelty search: a theoretical perspec- tive. In Proceedings of the Genetic and Evolutionary Computation Conference, pages 99–106, 2019. [4] C. Finn, P. Abbeel, and S. Levine. Model-agnostic meta-learning for fast adaptation of deep networks. In International Conference on Machine Learning, pages 1126– 1135. PMLR, 2017.
[5] B. Lim, L. Grillotti, L. Bernasconi, and A. Cully. Dynamics-aware quality-diversity for efficient learning of skill repertoires. arXiv preprint arXiv:2109.08522, 2021.
[6] A. Salehi, A. Coninx, and S. Doncieux. Few-shot quality-diversity optimisation. arXiv preprint arXiv:2109.06826, 2021.

Subject: From images to social interactions understanding

Abstract:

Studies in human-human interaction have introduced the concept of F-formation (Kendon, 1990) that defines three zones: private, social and public. Participants during social interaction place themselves in certain spatial formation. They can face each other, be side by side… Their position and behavior such as body orientation, gaze behavior can indicate a great quantity of information; they can reveal information about their level of engagement, their focus of interest but also the quality of their relationship, their degree of intimacy, to name a few. Participants’ position and behavior evolve continuously to accommodate others’ behaviors and to obey to some socio-cultural norms.

Lately computational models have been designed to detect if people form a group and its formation based on proxemics and behaviors (Cabrera-Quiros et al, 2018). Further analysis can be pursued to characterize the dynamics of the social interaction between participants. Such models can then be used to drive the behaviors of robots when interacting with humans.

Internship Objectives:

The aim of this internship is to analyze group interaction and their evolution over time. We will rely on existing data (images and videos) of group interaction that have been annotated at different levels (activity, speaking, laughing, non-verbal behavior). We will first make use of the database MatchNMingle (Raman&Hung,19).

Several steps are foreseen:

1. Perform a literature survey on F-formation detection, focusing in particular on videos and time evolution of the formations.

2. Perform tests using the method previously developed by V. Fortier last year.

3. Extend the model to videos.

4. Perform analysis of social actions such as predicting who will be the next speaker or the social relationship between interactants.

5. Depending on the project achievements, evaluate the results using virtual agents. MatchNMingle dataset: http://matchmakers.ewi.tudelft.nl/matchnmingle/pmwiki/index.php? n=Main.TheDataset

 

  • Supervisors: Isabelle Bloch / Catherine Pelachaud
  • Start date: February/april 2022
  • Duration: 6 months
  • Level of study required: Master 2
  • Location: ISIR (Institut des Systèmes Intelligents et de Robotique), Campus Pierre et Marie Curie, 4 place Jussieu, 75005 Paris.

 

Contact person:

  • Catherine Pelachaud et Isabelle Bloch
  • Email : catherine.pelachaud(at)upmc.fr ; isabelle.bloch(at)sorbonne-universite.fr
  • Send your application by email, with [internship subject] in the subject line, with CV, grades, names of referent and motivation letter (in pdf format)
  • Application deadline: January 2022

Download the internship offer

Subject: Developing computational models of perspective taking for human-robot interaction

Context:

Most collaborative robots will be dropped in human environment where they will have to accomplish missions. They will be engaged in interaction with human individuals concerned by these missions. Mutual understanding and coordination of goals, intentions, plans and actions are required to achieve these missions. In this work, we will focus on interaction involving robot motion generation such as arm movement or navigation and in particular we aim to generate legible motions.

Legible motions, also known as transparency through motion, are used by robots to communicate their intent to human observers [Dragan-2013]. By drawing inspiration from research on how humans interpret observed behaviors as goal-directed actions, the state-of- the-art approaches propose algorithms aiming to maximize human inference of a goal given to a robot motion. The approaches refer to “action-to-goal” interpretation of actions, which requires to explicitly taking into account the human observer in the loop.

In order to generate efficient legible motions, robots require a model of the human, which is called inverse human model. Building such models require perspective-taking abilities. Being able to place yourself “in someone else’s shoes” requires several sets of abilities such as the ability to understand the visuo-spatial experience of another agent; to infer her thoughts and beliefs; and to infer her emotions/feelings [Schilbach-2013]. These abilities are referred to as spatial, cognitive and affective perspective-taking, respectively [Hamilton-2014] and are investigated by means of experimental protocols that are now extensively used in cognitive sciences (e.g., [Sebanz-2006], [von Mohr-2020]. Despite the importance of these models, there is still no computational model able to compute on-line inverse models of perspective taking. In addition, it is not clear how humans will benefit from legible motions generated by such computational models.

Objectives:

The main objectives of this work are (i) to design experiments inspired by cognitive science allowing to evaluate perspective-taking during human-robot interactions (generation of robot motions), (ii) develop computational models able to predict perspective-taking preferences of human observers, (iii) develop metrics allowing to assess the benefit of computational inverse human models for the generation of legible robot motions.

This work will exploit methodologies and models of human-robot interaction, machine learning and cognitive science. The experiments can be carried out with the PEPPER or Franka Emika robots. Motion capture will be used to assess human reactions to robot motions.

The main steps are:

– Development of collaborative tasks such as cooperative drawing, teaching tasks requiring technical skills, or else object manipulation (see the “Give me the wrench scenario” [Trafton-2005] or spatial referring [Dogan-2020]),

– Design a study (including a control condition) to evaluate and collect multimodal data about perspective taking. Self-assessment, scales and questionnaires will be used to qualify human preferences,

– Development of machine learning based models able to detect human preferences,

– Experimental evaluation of models for the generation of legible motions.

Skills: Python, Machine Learning, Robotics, and Cognitive Science

 

  • Supervisors: Mohamed CHETOUANI,  Malika AUVRAY
  • Duration: 5/6 months
  • Level of study required: Master 2 / Engineering school
  • Location: ISIR (Institut des Systèmes Intelligents et de Robotique), Campus Pierre et Marie Curie, 4 place Jussieu, 75005 Paris.

 

Contact person:

  • Mohamed Chetouani
  • Tel : +33 1 44 27 08
  • Email : mohamed.chetouani(at)sorbonne-universite.fr
  • Send your application by email, with [internship subject] in the subject line, a CV and a cover letter.
  • Application deadline: 10/01/2022

Download the internship offer

Subject: Explainable Multimodal Machine Learning for TECH-TOYS: application to Precision Medicine models of developmental disorders

Context:

Neurodevelopmental disorders (NDDs) are a group of frequent (1/10 children) sensori-motor, cognitive, communication, learning, behavioral disorders of multifactorial aetiology, with onset early in life but life-long consequences. Despite advances in our understanding of aetiology, diagnosis and start of intervention are often late (many months after onset of first clinical signs) and not based on quantitative data.

The PIRoS team is engaged in the international TECH-TOYS project which aims to develop an innovative technological play setting (i.e., a gym equipped with a sensorized mat, a set of sensorized toys, wearable inertial movement units and cameras), suitable at infant’s natural environment, to provide easy-to-handle quantitative measurements of infant’s neurodevelopment and infant-caregiver interaction. Obtained data will provide a computational Precision Model for early detection of atypical features using digital biomarkers.

The study of digital biomarkers usually follows a retrospective methodology [Saint-Georges- 2010]. Several infants are engaged in a protocol, and after diagnosis (around 3 years) clinicians are able to define control and pathological groups. Computational models could be employed to characterize and predict developmental disorders based on the analysis of such longitudinal data [Saint-Georges-2011].

Objectives:

In this project, we investigate developmental disorders such as autism spectrum disorders. The data collected (during the 3 years) are then employed to develop predictive models of the pathology.

The objective of this work is to develop multimodal machine learning techniques to identify digital biomarkers grounded in Explainable Artificial Intelligence [Arrieta-2020; Wallkotter- 2021]. We will take advantage of a combination of previous big data already acquired (see CareToy project http://www.caretoy.eu) with new ones collected prospectively during the project.

This work will exploit methodologies and models of human-robot interaction and machine learning. We will use data collected in previous projects to develop predictive models of developmental disorders using multimodal signals. In addition to the model, we will work on the design of TECH-TOYS device. The work will include collaborations with TECH-TOYS partners in Italy, Turkey and Germany in order to address clinical, ethical, legal as well as scientific issues raised by the project.

The main steps are:

– Design a study for the prediction of developmental disorders based on CareToy data,

– Selection of relevant multimodal signals (sound, accelerometers, video),

– Development of machine learning based models able to predict developmental disorders,

– Define requirements for a future TECH-TOYS device.

Skills: Python, Machine Learning, and Robotics

 

  • Supervisor: Mohamed Chetouani
  • Duration: 5/6 months
  • Level of study required: Master 2 / Engineering school
  • Location: ISIR (Institut des Systèmes Intelligents et de Robotique), Campus Pierre et Marie Curie, 4 place Jussieu, 75005 Paris.

 

Contact person:

  • Mohamed Chetouani
  • Tel : +33 1 44 27 63 08
  • Email : mohamed.chetouani(at)sorbonne-universite.fr
  • Send your application by email, with [internship subject] in the subject line, a CV and a cover letter.
  • Application deadline: 15/12/2021

Download the internship offer

PhD offers

Thesis topic: Optical microrobots for interactive manipulation of biological samples

Abstract: This thesis aims at developing a new scientific instrument for applications in experimental biology, in particular for the manipulation, characterization and analysis of objects such as isolated cells, neurons, or intracellular organs. Using the principle of optical tweezers, laser beams are controlled to act directly on samples, or to actuate remote-controlled microrobots. These microrobots will be able to integrate analysis capabilities and bio-active sensors allowing a quick feedback to the operator. This is a new technology capable of supporting and considerably accelerating several studies in biology. Collaborations are started with teams from Institut Curie and Pasteur around cancer and intracellular mechanisms.

General description of the project: The optical tweezers are a technique allowing to manipulate microscopic objects by using a focused laser beam. They allow to act on samples in solution by a non-contact action. ISIR has developed a robotic laser trap system able to manipulate samples on 3 dimensions while measuring the interaction forces in real time. Nevertheless, the difficulty of handling these devices remains an important step to overcome, especially when it concerns objects outside the image plane.

The current performances of the system show that it is possible to trap and move simultaneously several particles with an effort resolution close to 10pN (Fig. A and B). Using these principles, optical microrobots have been realized. Activated by lasers, these ‘optobots’ (Fig. C) of a few micrometers in size, will be used to perform operations on biological samples, such as mechanical characterization, interaction measurement, genetic injection and electrical analysis. However, achieving such high performance has been at the expense of simplicity of use. This is mainly due to the design of the optical path and the complex control laws used.

The objective of this project is first to develop applications in experimental biology to demonstrate the advantages of this system and to impose it as a new scientific instrument. A collaboration is established with teams from the Pasteur Institute and the Curie Institute to exploit these possibilities in cancer research and studies on intracellular mechanisms. In this context, it is necessary to optimize interactivity so that the user is able to plan complex trajectories to trap and move objects, automate operations and collect results. We will also be interested in the modalities of Human/Machine interaction: dedicated haptic interfaces, notably among the previous achievements of the lab such as the ‘FishTank’, are promising candidates to develop a chain of cross-scale, multi-modal interaction.

Scientific theme: The main scientific theme is microrobotics, with strong support from physics and optics. The problems of object positioning and control in 6D with microscopic resolution and precision (nanometers and picoNewtons) are at the heart of the work. From an interaction point of view, existing solutions are generally difficult for the user to grasp and HMI approaches are an original way to achieve this. The user’s immersion is indeed an asset to free himself from complex control laws and planning systems. In the same way, the use of high performance integrated sensors is an asset concerning the final precision reached by the system.

Expected results, challenges and perspectives: The experience of ISIR in handling systems and in human-machine interaction allows us to envisage very promising perceptives and spin-offs. Such an achievement has never been done before and we are confident that it would be a major contribution to the use of optical tweezers. At the end of the project, applications in biology such as the manipulation of intracellular organs. These will be made possible with the collaboration of research teams in experimental biology.

This thesis is part of an industrial maturation process to create an innovative instrument in the field of life sciences, supported by the SATT and the Ile-de-France region. The perspectives concern an exploitation of the generated knowledge to accelerate research in biology. The creation of a start-up is also envisaged to valorize the results.

  • Thesis director: Sinan Haliyo
  • Possible co-supervision: Stéphane Régnier
  • Collaborations within the framework of the thesis: Institut Pasteur, Institut Curie
  • Location: Isir (Institut des Systèmes Intelligents et de Robotique), 4 Place Jussieu 75005, Paris, the Multi-Scale Interactions team
  • Contact: Sinan Haliyo ; sinan.haliyo@isir.upmc.fr ; Send your application by email, with [Thesis: Optical microrobots for interactive manipulation of biological samples] in the subject line, a CV and a cover letter.

Download the offer