Exploration of unsupervised approaches for modeling the environment from RADAR data
Radar technologies have gained significant interest in recent years, particularly with the emergence of MIMO radars and "Imaging Radars 4D". This new generation of radar offers both opportunities and challenges for the development of perception algorithms. Traditional algorithms such as FFT, CFAR, and DOA are effective for detecting moving targets, but the generated point clouds are still too sparse for precise environment model. This is a critical issue for autonomous vehicles and robotics.
This thesis proposes to explore unsupervised Machine Learning techniques to improve environment model from radar data. The objective is to produce a richer model of the environment to enhance data density and scene description, while controlling computational costs for real-time computing. The thesis will address the question of which types of radar data are best suited as inputs for algorithms and for representing the environment. The candidate will need to explore non-supervised algorithmic solutions and seek computational optimizations to make these solutions compatible with real-time execution.
Ultimately, these solutions must be designed to be embedded as close as possible to the sensor, allowing them to be executed on constrained targets.
Power and data transmission via an acoustic link for closed metallic environments
This thesis focuses on the transmission of power and data through metal walls using acoustic waves. Ultimately, this technology will be used to power, read and control systems located in areas enclosed in metal, such as pressure vessels, ship hulls and submarines.
Because electromagnetic waves are absorbed by metal, acoustic waves are needed to communicate data or power through metal walls. These are generated by piezoelectric transducers bonded to either side of the wall. The acoustic waves are poorly attenuated by the metal, resulting in numerous reflections and multiple paths.
The aim of the thesis will be to develop a robust demonstrator of this technology, enabling the remote powering and communication of acoustic data through metal walls. This work will be based on advanced modelling of the acoustic channel in order to optimise the performance of the power and data transmission device. It will also involve developing innovative electronic building blocks to determine and maintain an optimum power transmission frequency, impacted by environmental conditions and typically by temperature.
The goal of this thesis will be the development and implementation of a communication system embedded in an FPGA and/or microcontroller in order to send sensor data through a metal wall of variable thickness. The limitations due to the imperfections of the channel and the electronics will lead to the invention of a large number of compensation methods and systems in the digital and/or analogue domain. Work will also have to be carried out on the choice of piezoelectric transducers and the characterisation of the channel, in conjunction with the acoustic wave activities of the laboratory working on the transmission of acoustic power.