Scientific direction Development of key enabling technologies
Transfer of knowledge to industry

PhD : selection by topics

See all positions [+]

Human-in-the-Loop Learning and Adaptation under Uncertain and Unpredictable Situations in AI-based Autonomous Systems

Département Ingénierie Logiciels et Systèmes (LIST)

Labo.conception des systèmes embarqués et autonomes

01-10-2020

SL-DRT-20-1108

huascar.espinoza@cea.fr

Artificial intelligence & Data intelligence (.pdf)

Autonomous systems are evolving towards self-adaptive systems, being boosted by Artificial Intelligence (AI) techniques such as Machine/Deep Learning (M/DL). The emergence of autonomy means that software has to operate in an open and highly dynamic world, being capable of adapting themselves autonomously at run time to new environment conditions or unpredictable situations. In particular, this thesis aims to explore the combination of the capabilities of humans and algorithms to detect uncertainty regions and avoid dangerous situations in the real world and transfer control between a machine and a human (or to the safest agent). Learning-enabled systems based on deep learning are first trained in simulation environments before deploying them in the real world. While simulators are providing increasingly realistic training environments, there is always a gap between simulation and training, because training data does not capture some feature spaces and the model does not learn about them due to the incompleteness of the simulator to reflect the complexity of the real world. Our goal is to find methods for detecting unknown unknowns by combining simulation training with human input from demonstration data.

Download the offer (.zip)

High efficiency photovoltaic modules developement for building applications

Département des Technologies Solaires (LITEN)

Laboratoire Modules Photovoltaïques Silicium

01-10-2020

SL-DRT-20-1116

bertrand.chambion@cea.fr

Solar energy for energy transition (.pdf)

Performance of photovoltaic (PV) modules has not stopped to evolve in recent years to reach higher values at 20%. This is possible by a significant effort focussed on the architecture solar cells through gains in light absorption and better collection of photo-generated charges. In contrast, the module packaging and module structure remain similar as previous module structure. On one hand, these modules have been developed in order to work in a standard outdoor PV farm configuration. On the other hand, optimization and development are carried out under standard conditions where the temperature is set at 25 ° C. For Building Integrated (BIPV) applications, it can dramatically decrease their performance. This is related to the urban environment and local microclimate conditions (temperature, surrounding diffuse radiation), orientation and the tilt of the components. In addition, non-optimized integration conditions has a direct consequence and could increase the module temperature, making the thermal dependence on yield (estimated at -0.4% per degree) much more sensitive than in standard application. In addition, BIPV poses other issues related to the architectural aspects of the building: The quality of materials and their colours must match the environment, especially for old buildings. The purpose of this thesis project is to develop integrated PV modules prototypes, optimized for BIPV application, in accordance to the following steps: - State of the art on BIPV applications, materials, light management, thermal and thermomechanical modelling tools - Multi-scale modelling (cell, module, building, town) to understand the PV system thermal behaviour and performances consequences. - PV modules prototypes definition and realization on CEA INES Lab. - Outdoor ageing test and performances monitoring, comparison to standard PV solutions

Download the offer (.zip)

Reliable metal organic frameworks (MOF) and derivatives for radioactive gas detection

Département Métrologie Instrumentation et Information (LIST)

Laboratoire Capteurs et Architectures Electroniques

01-10-2020

SL-DRT-20-1117

guillaume.bertrand@cea.fr

This PhD position will be carried out in the frame of the SPARTE European project, which aims at developing porous scintillators in order to detect radioactive gases. The main goal of this PhD thesis is to explore the rich family of Metal Organic Frameworks (MOFs), and specifically the ones with fluorescent organic building blocks. The study will follow good preliminary results that our team got with Zinc-based MOFs and MOFs derivatives. We would like to extend the scope of these good results to other porous frameworks. Synthesis of new fluorescent ligands will also be considered. Photophysical properties of these materials will also be explored as well as their ability to concentrate and detect radioactive gases. This last measurement will be performed on a unique in the world labmade radioactive-gas bench. The PhD candidate is expected to perform inorganic solvothermal synthesis as well as photophysical characterization. He/she will also work on the radioactive gas bench set up. We wish to recruit somebody with a strong background in inorganic or organic chemistry and an open mind towards material-science technics. This thesis will be performed in two different laboratories, with two technical supervisers (Guillaume BERTRAND - LCAE and Benoît SABOT - LNHB/LMA) and a PhD director (Matthieu HAMEL - LCAE).

Download the offer (.zip)

Power Efficient AI-based IoT Physical Layer

Département Systèmes (LETI)

Laboratoire Communication des Objets Intelligents

01-10-2020

SL-DRT-20-1122

valerian.mannoni@cea.fr

Communication networks, IOT, radiofrequencies and antennas (.pdf)

The Internet of Things (IoT) is now a reality: more than 9 billion objects are already connected, 25 billion will be by 2025. To be effective, radio communication systems for the IoT must have a large coverage (operating at very low sensitivity levels) while having low energy consumption. In addition, the usual mode of operation of its systems is to communicate small messages sporadically and mainly uplink. The combination of all these constraints has led to the emergence of radio systems specific to the IoT such as LoRa, SigFox or NB-IoT and its current developments in 5G. However, it has been shown that these technologies do not simultaneously meet the needs of the IoT and that it is necessary to propose a new physical layer capable of addressing all the contradictory requirements of the IoT: - Low sensitivity (good performance at low spectral efficiency) for short messages (short error correcting codes and associated decoders). - Low power consumption (constant envelope waveform and high throughput). Limited and optimized signaling overhead. To achieve these objectives, algorithms based on Artificial Intelligence (AI) will be considered, in particular at the receiver/decoder level (embedded AI).

Download the offer (.zip)

DC/DC converter based on piezoelectric material and adiabatic power transfer

Département Systèmes (LETI)

Laboratoire Electronique Energie et Puissance

01-10-2020

SL-DRT-20-1148

ghislain.despesse@cea.fr

Energy efficiency for smart buildings, electrical mobility and industrial processes (.pdf)

The aim of this thesis is to design high-efficiency power converters based on resonating piezoelectric transducers. A large part of the work is to develop the electrical cycle able to energetically maintain the piezoelectric resonator in resonance and ensure zero-voltage switching, for electrical energy transfer from the source to the piezoelectric resonator or from the piezoelectric resonator to the output, in order to minimize the losses. An electronic power management circuit will be designed to enable this ideal energetic cycle. This electronic circuit will include several regulation loops to ensure the system stability and regulate the electrical output power. Finally, a study of the piezoelectric transducer size reduction will be done in view of a MEMS (Micro Electro Mechanical System) integration.

Download the offer (.zip)

173 Results found (Page 29 of 29)
first   previous  25 - 26 - 27 - 28 - 29

See all positions