AI and dynamical systems: new paradigms for control and robots

Description

The basic objective of this chair is to investigate ways by which the powerful tools of IA can be used to extend the domain of applicability and the performance of dynamical systems in order to handle complex relationships that are not easily modeled by knowledge-based equations. More specifically, we develop new paradigms and/or methodologies for control systems and robotics with AI methods. Motivated by recent improvements of classic control algorithms by encapsulating AI methods and algorithms, we tackle fundamental issues in control systems and robotic. By doing so, we will overcome the limitations of classical control theory by adding the AI in the loop, and by using AI in the dynamics. Moreover, for many problems, such as calibration of sensors and actuators, modeling in presence of uncertainty, and optimal uncertainty-aware control, AI methods have been seen as necessary paradigms to get adequate results. In the final objective of this chair, we will develop dynamic methodologies in AI methods and algorithms that are sometimes operate on the basis of static relationships.

Activities

This chair support research activities on control and robotic dynamical systems and on AI, in particular by providing grants to PhD students and post-doc positions. It brings together researchers and students with supports from Univ. Grenoble Alpes, Cnrs and Inria (at Gipsa-lab, Inria Grenoble and Lig), and industrial partners (Schneider and Sysnav among others).

Chair events

Meetings are organized inside of this chair with internal presentations, and scientific discussions. The members attend international conferences and are also often invited to give seminars and lectures dedicated to their activity related to the chair (as in the IEEE Columbian Control Conference, and the 14th ESA Workshop on Avionics, Data, Control and Software Systems).

Scientific publications

Lastest publication:

All scientific publications are included in the collection MIAI in HAL. They include :
  • Makia Zmitri, Hassen Fourati, Christophe Prieur. Improving Inertial Velocity Estimation Through Magnetic Field Gradient-based Extended Kalman Filter. IPIN 2019 - International Conference on Indoor Positioning and Indoor Navigation, Sep 2019, Pisa, Italy. pp.1-7. ?hal-02297850?
  • Nicolas Vanspranghe, Francesco Ferrante, Christophe Prieur. Control of a Wave Equation with a Dynamic Boundary Condition. 59th Conference on Decision and Control, 2020, Jeju Island, South Korea. ?hal-02987252?
  • David Collet, Mazen Alamir, Domenico Di Domenico, Guillaume Sabiron. Behavioral cloning for fatigue-oriented wind turbine optimal predictive control. International Symposium on Mathematical Theory of Networks and Systems, Aug 2020, Cambridge, United Kingdom. ?hal-02610805?
  • Philip Scales, Olivier Aycard, Véronique Auberge. Studying Navigation as a Form of Interaction: a Design Approach for Social Robot Navigation Methods. International Conference on Robotics and Automation (ICRA 2020), 2020, Paris, France. ?hal-02541820?