DFL-TORO: A One-Shot Demonstration Framework for Learning Time-Optimal Robotic Manufacturing Tasks

This paper presents DFL-TORO, a novel Demonstration Framework for Learning Time-Optimal Robotic tasks via One-shot kinesthetic demonstration. It aims at optimizing the process of Learning from Demonstration (LfD), applied in the manufacturing sector. As the effectiveness of LfD is challenged by the...

Full description

Saved in:
Bibliographic Details
Main Authors: Alireza Barekatain, Hamed Habibi, Holger Voos
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10735215/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850198701152141312
author Alireza Barekatain
Hamed Habibi
Holger Voos
author_facet Alireza Barekatain
Hamed Habibi
Holger Voos
author_sort Alireza Barekatain
collection DOAJ
description This paper presents DFL-TORO, a novel Demonstration Framework for Learning Time-Optimal Robotic tasks via One-shot kinesthetic demonstration. It aims at optimizing the process of Learning from Demonstration (LfD), applied in the manufacturing sector. As the effectiveness of LfD is challenged by the quality and efficiency of human demonstrations, our approach offers a streamlined method to intuitively capture task requirements from human teachers, by reducing the need for multiple demonstrations. Furthermore, we propose an optimization-based smoothing algorithm that ensures time-optimal and jerk-regulated demonstration trajectories, while also adhering to the robot’s kinematic constraints. The result is a significant reduction in noise, thereby boosting the robot’s operation efficiency. Evaluations using a Franka Emika Research 3 (FR3) robot for a variety of tasks further substantiate the efficacy of our framework, highlighting its potential to transform kinesthetic demonstrations in manufacturing environments. Moreover, we take our proposed framework into a real manufacturing setting operated by an ABB YuMi robot and showcase its positive impact on LfD outcomes by performing a case study via Dynamic Movement Primitives (DMPs).
format Article
id doaj-art-d2b5e820505245f8acece6f54457ea8e
institution OA Journals
issn 2169-3536
language English
publishDate 2024-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-d2b5e820505245f8acece6f54457ea8e2025-08-20T02:12:49ZengIEEEIEEE Access2169-35362024-01-011216116416118410.1109/ACCESS.2024.348661510735215DFL-TORO: A One-Shot Demonstration Framework for Learning Time-Optimal Robotic Manufacturing TasksAlireza Barekatain0https://orcid.org/0000-0001-5646-2675Hamed Habibi1https://orcid.org/0000-0002-7393-6235Holger Voos2https://orcid.org/0000-0002-9600-8386Interdisciplinary Centre for Security, Reliability and Trust (SnT), University of Luxembourg, Luxembourg City, LuxembourgInterdisciplinary Centre for Security, Reliability and Trust (SnT), University of Luxembourg, Luxembourg City, LuxembourgInterdisciplinary Centre for Security, Reliability and Trust (SnT), University of Luxembourg, Luxembourg City, LuxembourgThis paper presents DFL-TORO, a novel Demonstration Framework for Learning Time-Optimal Robotic tasks via One-shot kinesthetic demonstration. It aims at optimizing the process of Learning from Demonstration (LfD), applied in the manufacturing sector. As the effectiveness of LfD is challenged by the quality and efficiency of human demonstrations, our approach offers a streamlined method to intuitively capture task requirements from human teachers, by reducing the need for multiple demonstrations. Furthermore, we propose an optimization-based smoothing algorithm that ensures time-optimal and jerk-regulated demonstration trajectories, while also adhering to the robot’s kinematic constraints. The result is a significant reduction in noise, thereby boosting the robot’s operation efficiency. Evaluations using a Franka Emika Research 3 (FR3) robot for a variety of tasks further substantiate the efficacy of our framework, highlighting its potential to transform kinesthetic demonstrations in manufacturing environments. Moreover, we take our proposed framework into a real manufacturing setting operated by an ABB YuMi robot and showcase its positive impact on LfD outcomes by performing a case study via Dynamic Movement Primitives (DMPs).https://ieeexplore.ieee.org/document/10735215/ABB YuMiFranka Emika Research 3imitation learninglearning from demonstrationmanufacturing roboticsrobotic manipulator
spellingShingle Alireza Barekatain
Hamed Habibi
Holger Voos
DFL-TORO: A One-Shot Demonstration Framework for Learning Time-Optimal Robotic Manufacturing Tasks
IEEE Access
ABB YuMi
Franka Emika Research 3
imitation learning
learning from demonstration
manufacturing robotics
robotic manipulator
title DFL-TORO: A One-Shot Demonstration Framework for Learning Time-Optimal Robotic Manufacturing Tasks
title_full DFL-TORO: A One-Shot Demonstration Framework for Learning Time-Optimal Robotic Manufacturing Tasks
title_fullStr DFL-TORO: A One-Shot Demonstration Framework for Learning Time-Optimal Robotic Manufacturing Tasks
title_full_unstemmed DFL-TORO: A One-Shot Demonstration Framework for Learning Time-Optimal Robotic Manufacturing Tasks
title_short DFL-TORO: A One-Shot Demonstration Framework for Learning Time-Optimal Robotic Manufacturing Tasks
title_sort dfl toro a one shot demonstration framework for learning time optimal robotic manufacturing tasks
topic ABB YuMi
Franka Emika Research 3
imitation learning
learning from demonstration
manufacturing robotics
robotic manipulator
url https://ieeexplore.ieee.org/document/10735215/
work_keys_str_mv AT alirezabarekatain dfltoroaoneshotdemonstrationframeworkforlearningtimeoptimalroboticmanufacturingtasks
AT hamedhabibi dfltoroaoneshotdemonstrationframeworkforlearningtimeoptimalroboticmanufacturingtasks
AT holgervoos dfltoroaoneshotdemonstrationframeworkforlearningtimeoptimalroboticmanufacturingtasks