Advances in control and estimation of complex systems: living behavior and multistability
This Ph.D. thesis takes part in an interdisciplinary collaborative project (ANRWaQMoS) joining marine biology, electronics, and applied mathematics. This project’s primary goal is to develop an intelligent autonomous biosensor based on the measurement and interpretation of bivalve mollusks’ behavior...
Main Author: | |
---|---|
Other Authors: | , , , , , |
Format: | Doctoral or Postdoctoral Thesis |
Language: | English |
Published: |
HAL CCSD
2020
|
Subjects: | |
Online Access: | https://hal.inria.fr/tel-03991659 https://hal.inria.fr/tel-03991659/document https://hal.inria.fr/tel-03991659/file/2020LILUI057.pdf |
Summary: | This Ph.D. thesis takes part in an interdisciplinary collaborative project (ANRWaQMoS) joining marine biology, electronics, and applied mathematics. This project’s primary goal is to develop an intelligent autonomous biosensor based on the measurement and interpretation of bivalve mollusks’ behavioral responses from environmental stimulus. The principal application is remote coastal water quality surveillance and ecosystem change monitoring in sensitive areas due to pollutions or climate change consequences. The biosensor utilizes a high-frequency non invasive valvometry technology combined with a data acquisition system. In its turn, the data acquisition block includes a complex, intelligent tool, which aims to convert the behavioral responses of bivalve mollusks into a set of useful information for indirect ecological monitoring. The possibility of such a conversion comes from the fact that these animals are quite sensitive to their environmental changes. Moreover, characteristics of their reactions can be captured from the opening/closing movements of its valves, measured by the sensor. However, understanding, isolating, and connecting the information contained in the distance signals to the environmental or climate entrances form a significant challenge to the biosensor’s realization.Following this problematic, motivated by the climate change subject, in the first part of this thesis, we aim to provide a set of time-scaled populational behavioral variables obtained from the distance signal measured along several years of data acquisition in the Arctic region. These aggregated biologically meaningful variables can be related to bioclimatic ones, such as time scaled air and water surface temperature or their maximum and minimum variations. For this objective, a sequence of data processing tools was proposed, including an intelligent adaptive filter, based on advanced velocity estimation and the dynamic regressor extension and mixing method (DREM) with fixed-(finite-)time estimation approaches. The obtained ... |
---|