Postdoctoral position (M/F) experimentalist in swarm robotics and colloective learning

Job Description

Missions

The postdoctoral researcher will be responsible for developing and conducting the experimental component of a research project focused on the emergence of forms of collective intelligence in active robotic systems. The project's overall objective is to understand how a population of autonomous robots—interacting locally and without centralized control—can collectively learn behavioral rules adapted to a specific task and environment. Particular attention will be paid to the coupling between the system's physical dynamics (movement, steric interactions, confinement, density, assembly) and the dynamics of learning and information transmission among agents. The researcher will develop quantitative experiments to explore the feedback loop linking physical organization, interactions, social learning, and collective performance. In particular, they will help identify the conditions under which physical interactions act as a barrier to collective learning or, conversely, can be leveraged as a resource.

Activity

The postdoctoral researcher will be responsible for designing, conducting, and analyzing experiments involving populations of small, programmable mobile robots. Their activities will include: extending the capabilities of the existing swarm robotics platform, particularly through the potential development of new robotic features to explore scenarios involving adhesion, assembly, or collective deformation; implementing decentralized social learning protocols where agents can modify and exchange behavioral policies based on their experience; designing experimental tasks to distinguish between individual and truly collective performance—such as gradient navigation, exploration, adaptation to changing environments, aggregation, transport, or obstacle crossing; systematically studying how factors like density, confinement, boundaries, steric interactions, and interaction network structure affect learning dynamics; acquiring and processing quantitative data, including trajectories, neighborhoods and contacts, inter-robot communication, internal states and individual policies, as well as performance metrics and learning times; characterizing collective properties such as policy diversity, fixation rates, spatial mixing, interaction network turnover, robustness to perturbations, and adaptability to new tasks; comparing experimental results with numerical simulations and theoretical models developed within the project; and participating in the supervision of students involved in the project, attending consortium meetings, and disseminating findings through publications and presentations at international conferences. A significant portion of the work will involve designing experiments that are simple enough to allow for quantitative physical analysis while retaining the essential elements of agent learning and autonomy.

Your Profil

Skills

The position is aimed at a candidate with a doctorate in physics, complex systems, active matter, or in a related field. The following skills will be particularly appreciated: solid experience in experimental physics and a strong taste for designing quantitative experiments; ability to develop, modify and instrument experimental devices; very good programming skills, ideally in Python and/or C/C++; experience in acquiring and processing experimental data; knowledge of mobile robotics, embedded systems, electronics or communication protocols between agents, appreciated but not essential; interest in active matter, dynamic systems, non-equilibrium statistical physics, complex systems or collective robotics; interest in decentralized learning methods, evolutionary learning or collective intelligence; ability to work at the interface between experiment, simulation and theory; scientific autonomy, experimental creativity and ability to work within an interdisciplinary team; good command of written and oral scientific English. Prior experience in robotics is not essential if the candidate has a solid experimental culture and a strong interest in instrumental development and programming.

Your Work Environment

The postdoctoral researcher will be hosted at the Gulliver Laboratory at ESPCI Paris – PSL, under the supervision of Olivier Dauchot. The Gulliver Laboratory conducts highly interdisciplinary research at the interface of statistical physics, soft matter, active matter, and complex systems. The project will benefit from an environment combining experiments, modeling, numerical simulations, and theoretical developments. The postdoctoral researcher will work in close collaboration with researchers involved in the project on aspects of active matter, swarm robotics, social learning, and non-equilibrium statistical physics. They will have access to an existing robotic platform—serving as the starting point for experiments—as well as simulation and analysis tools developed in parallel within the consortium. The project thus offers the opportunity to contribute to the emergence of a field at the interface of active matter, robotic matter, and collective intelligence, with the aim of understanding the conditions under which a physical collective can not only self-organize but also learn to modify the very rules of its organization.

Constraints and risks

The position entails no specific occupational constraints or risks beyond those associated with routine laboratory experimental work. The experiments primarily involve small robotic devices, low-voltage electronics, programming, and data acquisition. Standard laboratory safety rules apply.

Compensation and benefits

Compensation

€3,131 gross per month

Annual leave and RTT

44 jours

Remote Working practice and compensation

Pratique et indemnisation du TT

Transport

Prise en charge à 75% du coût et forfait mobilité durable jusqu’à 300€

About the offer

About the CNRS

The CNRS is a major player in fundamental research on a global scale. The CNRS is the only French organization active in all scientific fields. Its unique position as a multi-specialist allows it to bring together different disciplines to address the most important challenges of the contemporary world, in connection with the actors of change.

CNRS

The research professions