High mobility mobile manipulator control in a dynamic context

The development of mobile manipulators capable of adapting to new conditions is a major step forward in the development of new means of production, whether for industrial or agricultural applications. Such technologies enable repetitive tasks to be carried out with precision and without the constraints of limited workspace. Nevertheless, the efficiency of such robots depends on their adaptation to the variability of the evolutionary context and the task to be performed. This thesis therefore proposes to design mechanisms for adapting the sensory-motor behaviors of this type of robot, in order to ensure that their actions are appropriate to the situation. It envisages extending the reconfiguration capabilities of perception and control approaches through the contribution of Artificial Intelligence, here understood in the sense of deep learning. The aim is to develop new decision-making architectures capable of optimizing robotic behaviors for mobile handling in changing contexts (notably indoor-outdoor), and for carrying out a range of precision tasks.

Dynamics of a Superheated Steam Cycle in Flexible Electrical-Heat Power Generation with Grid Service: Stability Analysis and Associated Control Using Advanced Methods

The objective of the thesis is to develop a methodology for stability analysis and control, tailored to the unique characteristics of the energy conversion cycle of a Small Modular Reactor operating in cogeneration with grid service. The doctoral student will first focus on the control of the steam generator (SG) for a given technology, first with the development of a 1D dynamic model of the SG and then its flow stability analysis. Investigation of the control-command strategies will be progressively extended by coupling the GV to the energy conversion system and primary loop: the appropriate conditions for frequency control or load follow operation will be sought, in both electrical and cogeneration production modes. The approach should make it possible to quantify the gain that could be expected from advanced control techniques using machine learning in particular. Different mdeling tools will be used, including Dymola, Matlab together with some internal CEA or free tools.
The thesis benefits from a dual academic supervision. It will take place in the Innovative Nuclear System Study R&D Unit of the Nuclear Systems Research Institute at CEA Cadarache. The doctoral student will develop skills in energy conversion, thermohydraulics, control (strong component of the thesis) and artificial intelligence.

Learning Fine-Grained Dexterous Manipulation through Vision and Kinesthetic Observations

Fine-grained dexterous manipulation presents significant challenges for robots due to the need for precise object handling, coordination of contact forces, and utilization of visual observations. This research aims to address these challenges by investigating the integration of vision and kinesthetic sensors, sim2real techniques, and generalization through embodiment. The objective is to develop end-to-end algorithms and models that enable robots to manipulate objects with exceptional precision and adaptability. The research will focus on learning from large-scale data, transferring knowledge from simulations to real-world scenarios, and efficiently generalizing through low-shot fine-tuning.