Feedback

Faculté des Sciences appliquées
Faculté des Sciences appliquées
MASTER THESIS
VIEW 28 | DOWNLOAD 378

Master thesis : Deep Reinforcement Learning for Robotic Grasping

Download
Fares, Nicolas ULiège
Promotor(s) : Ernst, Damien ULiège ; Sacré, Pierre ULiège
Date of defense : 5-Sep-2022/6-Sep-2022 • Permalink : http://hdl.handle.net/2268.2/16288
Details
Title : Master thesis : Deep Reinforcement Learning for Robotic Grasping
Translated title : [fr] Apprentissage par renforcement profond pour la préhension robotique
Author : Fares, Nicolas ULiège
Date of defense  : 5-Sep-2022/6-Sep-2022
Advisor(s) : Ernst, Damien ULiège
Sacré, Pierre ULiège
Committee's member(s) : Wehenkel, Louis ULiège
Ewbank, Tom ULiège
Language : English
Number of pages : 83
Keywords : [en] Reinforcement Learning
[en] Robotic Grasping
[en] Deep Learning
Discipline(s) : Engineering, computing & technology > Computer science
Funders : Financement Win2Wal
Research unit : Montefiore
Name of the research project : IntegrIA
Target public : Researchers
Professionals of domain
Student
Institution(s) : Université de Liège, Liège, Belgique
Degree: Master : ingénieur civil en science des données, à finalité spécialisée
Faculty: Master thesis of the Faculté des Sciences appliquées

Abstract

[en] The development and deployment of robotic grasping systems in the industry help to improve the efficiency and productivity of one’s production lines.
Even though interesting for any industrial actor, those robotic systems require a significant upfront investment.
This significant investment is composed of two primary types of costs: hardware and software.
Thanks to recent developments in Deep Reinforcement Learning applied to robotic grasping through vision-based systems, IntegrIA is researching solutions that could reduce the software costs of robotic grasping applications focused on pick-and-place tasks.

Thus, this master’s thesis implements a state-of-the-art reinforcement learning algorithm named QT-Opt and aims to compare it with IntegrIA’s one.
Both online and offline learning versions of QT-Opt are developed, resulting in three training algorithms to compare across three training datasets.
Performances of resulting agents are quantitatively evaluated and qualitatively compared through metrics such as the normalised area under the success rate curve.

In the end, it is observed that this master thesis best agent trained on a dataset composed of 1,800 objects achieves a grasping success rate of 96.67% on previously unseen objects, against 97.32% for IntegrIA’s agent.
Even though it cannot outperform their implementation, it is interesting to observe that the best agent trained for this master’s thesis achieves the 96% success rate from the original paper while being powered with a fraction of its resources.


File(s)

Document(s)

File
Access Nicolas_Fares_Thesis.pdf
Description:
Size: 6.21 MB
Format: Adobe PDF

Annexe(s)

File
Access Nicolas_Fares_Abstract.pdf
Description:
Size: 81.16 kB
Format: Adobe PDF

Author

  • Fares, Nicolas ULiège Université de Liège > Master ingé. civ. sc. don. à . fin.

Promotor(s)

Committee's member(s)

  • Wehenkel, Louis ULiège Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Méthodes stochastiques
    ORBi View his publications on ORBi
  • Ewbank, Tom ULiège Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Méthodes stochastiques
    ORBi View his publications on ORBi
  • Total number of views 28
  • Total number of downloads 378










All documents available on MatheO are protected by copyright and subject to the usual rules for fair use.
The University of Liège does not guarantee the scientific quality of these students' works or the accuracy of all the information they contain.