Ph. D. Project
Title:
Machine learning and data fusion for environmental monitoring. Application to risk exposure prevention in working environments.
Dates:
2022/04/26 - 2025/04/25
Student:
Supervisor(s):
Other supervisor(s):
Philippe Duquenne (philippe.duquenne@inrs.fr)
Description:
Recent technological developments (sensor networks, geolocation, 3D scanners, etc.)open up new fields of application in environmental monitoring and
pollution exposure risk prevention. However, the analysis and representation of data acquired through these new technologies raise scientific questions that
must be addressed in order to fully benefit from technical progress. These issues mainly concern the creation and interpretation of spatio-temporal pollution
maps and the efficient fusion of data provided by sensors of different natures (gas/particles), or by other acquisition systems (3D scanner, localization
system etc.).
The objective of the thesis is to develop an automatic method for the spatio-temporal mapping of pollutant concentrations as well as a methodology for the
assessment of workers' exposure risks, using data acquired by different types of measuring instruments deployed in the working environment. This project
fits naturally into the current digital transformation of manufacturing/production and related industries, commonly referred to as "Industry 4.0".
pollution exposure risk prevention. However, the analysis and representation of data acquired through these new technologies raise scientific questions that
must be addressed in order to fully benefit from technical progress. These issues mainly concern the creation and interpretation of spatio-temporal pollution
maps and the efficient fusion of data provided by sensors of different natures (gas/particles), or by other acquisition systems (3D scanner, localization
system etc.).
The objective of the thesis is to develop an automatic method for the spatio-temporal mapping of pollutant concentrations as well as a methodology for the
assessment of workers' exposure risks, using data acquired by different types of measuring instruments deployed in the working environment. This project
fits naturally into the current digital transformation of manufacturing/production and related industries, commonly referred to as "Industry 4.0".
Keywords:
spatiotemporal data analysis, machine learning, tensors, data fusion
Department(s):
| Biology, Signals and Systems in Cancer and Neuroscience |
