Author: Claire DUNE

Clémentin Boittiaux PHD Thesis Defense

The Doctoral Studies Office is pleased to inform you that

Mr Clémentin BOITTIAUX

Doctoral student at the COSMER laboratory

to Doctoral School 548 “Sea & Sciences”

will submit his thesis in order to obtain the degree of Doctor

under the supervision of

Vincent HUGEL, University Professor, University of Toulon (France)

&

Aurélien ARNAUBEC, PhD, Ifremer (France), Co-supervisor

Claire DUNE, Associate Professor, University of Toulon (France), Co-supervisor

Ricard MARXER, University Professor, University of Toulon (France), Co-supervisor

Discipline: Automatics, Signal, Productics, Robotics

on the theme

“Visual localization for long-term deep-sea monitoring”.

Thursday, December 14, 2023 at 10:00 a.m.
At Université de Toulon – Campus La Garde Bâtiment Y1 – Amphi Y.008

before a jury composed of :

 

  1. Ms Marie-Odile BERGER, Director of Research, INRIA Nancy Grand-Est (France), Rapporteur
  2. Vincent LEPETIT, University Professor, Ecole Nationale des Ponts et Chaussées (France), Rapporteur
  3. Cédric DEMONCEAUX,Professor, University of Burgundy (France), Examiner
  4. Torsten SATTLER,Research Fellow, Czech Institute of informatics Robotics and Cybernetics (Czech Republic), Examiner
  5. Aurélien ARNAUBEC,PhD, Ifremer (France), Co-supervisor
  6. Ms Claire DUNE, Associate Professor, University of Toulon (France), Co-supervisor
  7. Ricard MARXER,Professeur des Universités, Université de Toulon (France), Co-supervisor
  8. Vincent HUGEL,Professor, University of Toulon (France), Thesis supervisor

 

Visual localization for deep-sea long-term monitoring

Abstract :

This thesis explores the challenge of localizing underwater vehicles within previously explored environments in long-term scenarios, where significant scene appearance changes may have occurred. Typically, underwater vehicle positioning relies on fusing measurements from acoustic and inertial sensors. While these sensors deliver precise relative pose estimations, their absolute position estimates exhibit notable biases, resulting in position offsets spanning tens of meters between different dives. This limitation impedes the practical use of autonomous underwater vehicles for tasks requiring high precision, like mapping a precise area of interest. In response, this thesis investigates the use of visual observations made by underwater vehicles to enhance absolute positioning accuracy. The underwater environment introduces unique sources of variability absent in terrestrial environments. Consequently, the first contribution of this thesis is a novel dataset designed for benchmarking long-term visual localization algorithms in deep-sea conditions. Another obstacle inherent to underwater images is that their suffer from low contrast and loss of colors because of light propagation in the water medium. To address this issue, the second contribution of this work introduces two underwater color restoration methods, specifically designed to mitigate these phenomena and recover clear images. Independent of the underwater environment, the third contribution of this thesis is a novel loss function tailored for camera pose regression within the context of deep learning applications. This is an important aspect to consider when training visual localization networks. Finally, this thesis concludes with a benchmark of several visual localization methods on the proposed dataset. The obtained results show that applying our underwater color restoration method improves visual localization performance. This work also identifies the major problem encountered by visual localization methods on the proposed underwater dataset, and presents an approach to improve the accuracy of visual localization techniques by making the most of a limited size dataset.

Keywords: visual localization, deep-sea mapping.

 

 

 

PhD Defense Juliette Drupt

The Doctoral Studies Office is pleased to inform you that

Ms Juliette DRUPT

Doctoral student at the COSMER laboratory, part of the Doctoral School 548 “Mer & Sciences” (France)

will submit her thesis in order to obtain the degree of Doctor

under the supervision of

Vincent HUGEL,Professor, University of Toulon (France), Thesis Director

&

Mme Claire DUNE, Maitre de conférences, Université de Toulon (France), Thesis co-supervisor

Andrew COMPORT, Research Fellow, CNRS (France), Thesis co-supervisor

Discipline: Automatics, Signal, Productics, Robotics

on the theme

“Localization of an underwater robot chain“.

Wednesday, November 29, 2023 at 10:00 a.m.

At Université de Toulon – Campus La Garde – Bâtiment M – Amphi M.01

before a jury composed of :

  1. David FILLIAT, Professeur des Universités, ENSTA Paris (France), Rapporteur,
  2. Luc JAULIN, Professeur des Universités, ENSTA Bretagne (France), rapporteur,
  3. Vincent CREUZE, Professeur des Universités, Université de Montpellier (France), Examinateur,
  4. Juan TARDOS, Professeur, Université de Saragosse (Espagne), Examinateur,
  5. Maxime FERRERA, Docteur, ingénieur, IFREMER (France), Examinateur,
  6. Mme Claire DUNE, Maître de Conférences, Université de Toulon (France), co-encadrante de thèse
  7. Andrew COMPORT, Chargé de Recherches HDR, CNRS-I3S (France), Co-encadrant de thèse
  8. Vincent HUGEL, Professeur des Universités, Université de Toulon (France), Directeur de thèse

Abstract

Because real-time wireless communications are impossible underwater, underwater remotely operated vehicles (ROVs)  are connected to a surface station by a physical link. Recent works investigate limiting the mechanical effects exerted by this cable by using intermediary ROVs  to control its shape.  This configuration is called an underwater robot chain and is the focus of the current thesis, which investigates its self-localization using embedded sensors.

While ROV localization with respect to its environment can be estimated from its embedded sensors, the cable can be taken as an advantage to localize the ROV at its end point given a knowledge of the cable’s 3D state. Two complementary approaches are studied: the proprioceptive localization of the chain based on an estimation of the 3D state of its cable parts using inertial measurements, and exteroceptive, multi-agent localization with respect to the environment using visual simultaneous localization and mapping techniques.

Keywords: underwater robotics, deformable objects, VSLAM.

PHD Defense of Lewis ANDURAND

The Doctoral Studies Office is pleased to inform you that

Mr. Lewis ANDURAND

Doctoral student at the COSMER laboratory

laboratory, attached to Doctoral School 548 “Mer & Sciences” (France)

will submit his dissertation in view of obtaining the Grade de Docteur

under the supervision of

Vincent HUGEL, Professor, University of Toulon (France)

Co-supervised by

Sébastien CAMPOCASSO, Associate Professor, University of Toulon (France)

&

Matthieu MUSEAU, Associate Professor, Grenoble-Alpes University (France)

Discipline: Solid Mechanics, Mechanical Engineering, Productics, Transport and Civil Engineering

Specialty: Additive Manufacturing

on the theme

“Development of a versatile trajectory generation method for the production of multi-axis DED parts from faceted surfaces”.

Thursday, August 31, 2023 at 09:30 a.m.
At Université de Toulon – Campus La Garde – Bâtiment M – Amphi M.01

before a jury composed of :

  1. Jean-Philippe PERNOT, University Professor, École Nationale Supérieure d’Arts et Métiers, Rapporteur
  2. Stéphane SEGONDS, Professor, University of Toulouse III, Rapporteur
  3. François VILLENEUVE, Professeur des Universités, Université Grenoble-Alpes, Examiner
  4. Sébastien CAMPOCASSO, Senior Lecturer, University of Toulon, Supervisor
  5. Matthieu MUSEAU, Senior Lecturer, Grenoble-Alpes University, Supervisor
  6. Vincent HUGEL, Professor, University of Toulon, Thesis Director

 

Development of a Versatile Path Generation Method for the Manufacturing of Parts by Multi-Axis DED Processes from Meshed Surfaces

Abstract:

Additive manufacturing is a category of processes that allows the production of mechanical parts by the adding of material. Directed Energy Deposition (DED) processes can be combined with multi-axis robots and are a promising option to obtain parts with complex structures. However, the path generation methods and the machine structures used remain an issue. With innovations in these areas, the industrial possibilities would increase tenfold.

This thesis presents a numerical and systematic path generation method based on meshed surfaces and adapted to DED processes. The method was validated through simulations on minimal triply periodic surfaces and allows the creation of a first deposition path that meets the distance constraint between the part and the tool. This first path can be combined with region prioritization feedback to obtain a final path adapted to the physical warnings provided by the robot, the manufacturing material and the tool.

Keywords : Path Generation, Additive Manufacturing, Directed Energy Deposition, Computer Aided Manufacturing

 

Martin Filliung

Posted on by

Doctoral student Funding: French Research and Industry Ministry Grant 2022-2025 E-mail: Martin-filliung (at) etud.univ-tln.fr Office: Building M, University of Toulon, Campus La Garde.   Biography   Research subject Thesis director: Read more →

Cristiam Lasso

Posted on by

PhD student Mail : classope134 (at) etud.univ-tln.fr Linkedin : Cristiam Javier Lasso Perdomo | LinkedIn Bio   Research Thesis topic: Architectural exploration method for an optimized helicopter structure obtained by Read more →

12-13-2021 PhD Defense of Ornella Tortorici

The Office of doctoral Studies is pleased to inform you that
Mrs Ornella TORTORICI
PhD candidate at COSMER & IM2NP laboratories with Doctoral School 548 « Sea & Sciences »
Under the supervision of M. Vincent HUGEL, Professor, Université de Toulon (France)
&

M. Hervé BARTHELEMY, Professor, University of Toulon (France), Codirecteur de thèse

supervised by M. Cédric ANTHIERENS, Assistant Professor, University of Toulon (France)

will defend her thesis to obtain the degree of Doctor

Discipline : « Automatic, Signal, Productics, Robotics »

on the thème

Design and automatic control of an instrumented umbilical for underwater robots
 

Monday December 13, 2021 at 1:30 pm

at University of Toulon – Campus La Garde – Amphi Building M

in front of the following jury

 

  • M. Benoit CLEMENT, Professor, ENSTA Bretagne, reviewer
  • M. Vincent CREUZE, Assistant professor, University of Montpellier , reviewer
  • M. Gérard POISSON, Professor, IUT de BOURGES, examiner
  • M. Cédric ANTHIERENS, Assistant professor, University of Toulon, co-supervisor
  • M. Hervé BARTHELEMY, Professor, University of Toulon, supervisor
  • M. Vincent HUGEL, Professor, University of Toulon, co-supervisor
  • Mme Elisabeth MURISASCO, Professor, University of Toulon, invited

Abstract

Remotely operated underwater robots (ROV) are connected by an umbilical to communicate with their control station on the surface. This umbilical has a safety advantage as a physical link to the robot but also impacts the system by transmitting forces and disturbances to the ROV, increasing the risk of snags and limiting its working range. This thesis focuses on the active management of umbilical connected to small ROVs in order to limit their impact on the system.

The forces transmitted by the cable to the ROV are studied as a function of different parameters thanks to the chain model and then thanks to the construction of a finite solids model under Matlab Simulink. A simple method of estimating the torsional stiffness and bending stiffness coefficients of the cable is proposed to parameterize these models.

A solution to control the cable distribution by a reel based on its instrumentation with a passive compliance system and a bending sensor is then proposed. This solution is evaluated through simulations of the complete system under Vortex and tank tests of the ROV with its cable instrumented and controlled by the reel. The experiments showed the feasibility of the automatic control of the cable at a reasonable length adapted to the movements of the ROV thanks to its instrumentation.

Key words : Underwater robotics, Mechatronic design, Active management of instrumented umbilical

Abstract

Design and control of actuated tether for underwater robots

Remotely operated vehicles in underwater (ROV) are connected by an umbilical to communicate with their control unit on the surface. This umbilical has a safety advantage as a physical link to the robot but also impacts the system by transmitting forces and disturbances to the ROV, thus increasing the risk of snags and limiting its working range. This thesis focuses on the active management of umbilicals connected to small ROVs in order to limit their impact on the system.

The forces transmitted by the cable to the ROV are studied as a function of different parameters using the catenary model and the construction of a finite solids model under Matlab Simulink. A simple method to estimate the torsional stiffness and bending stiffness coefficients of the cable is proposed in order to parameterize these models.

A solution to control the cable distribution by a winch using its instrumentation which is composed of a passive compliance system and a bending sensor is then proposed. This solution is evaluated through simulations of the complete system under Vortex and tank tests of the ROV with its instrumented cable controlled by the reel. The experiments showed the feasibility of the automatic control of the cable at a length adapted to the movements of the ROV thanks to its instrumentation.

Keywords: Underwater robotics, Mechatronic design, Active management of instrumented tether

Translated with www.DeepL.com/Translator (free version)