Augmented Virtuality Input Demonstration Refinement Improving Hybrid Manipulation Learning for Bin Picking
Blank A, Zikeli L, Reitelshöfer S, Karlidag E, Franke J (2023)
Publication Type: Conference contribution
Publication year: 2023
Journal
Publisher: Springer Science and Business Media Deutschland GmbH
Pages Range: 332-341
Conference Proceedings Title: Lecture Notes in Mechanical Engineering
Event location: Detroit, MI
ISBN: 9783031183256
DOI: 10.1007/978-3-031-18326-3_32
Abstract
Beyond conventional automated tasks, autonomous robot capabilities aside human cognitive skills are gaining importance in industrial applications. Although machine learning is a major enabler of autonomous robots, system adaptation remains challenging and time-consuming. The objective of this research work is to propose and evaluate an augmented virtuality-based input demonstration refinement method improving hybrid manipulation learning for industrial bin picking. To this end, deep reinforcement and imitation learning are combined to shorten required adaptation timespans to new components and changing scenarios. The method covers initial learning and dataset tuning during ramp-up as well as fault intervention and dataset refinement. For evaluation standard industrial components and systems serve within a real-world experimental bin picking setup utilizing an articulated robot. As part of the quantitative evaluation, the method is benchmarked against conventional learning methods. As a result, required annotation efforts for successful object grasping are reduced. Thereby, final grasping success rates are increased. Implementation samples are available on: https://github.com/FAU-FAPS/hybrid_manipulationlearning_unity3dros
Authors with CRIS profile
Andreas Blank
Lehrstuhl für Fertigungsautomatisierung und Produktionssystematik
Sebastian Reitelshöfer
Lehrstuhl für Fertigungsautomatisierung und Produktionssystematik
Jörg Franke
Lehrstuhl für Fertigungsautomatisierung und Produktionssystematik
How to cite
APA:
Blank, A., Zikeli, L., Reitelshöfer, S., Karlidag, E., & Franke, J. (2023). Augmented Virtuality Input Demonstration Refinement Improving Hybrid Manipulation Learning for Bin Picking. In Kyoung-Yun Kim, Leslie Monplaisir, Jeremy Rickli (Eds.), Lecture Notes in Mechanical Engineering (pp. 332-341). Detroit, MI, US: Springer Science and Business Media Deutschland GmbH.
MLA:
Blank, Andreas, et al. "Augmented Virtuality Input Demonstration Refinement Improving Hybrid Manipulation Learning for Bin Picking." Proceedings of the 31st International Conference on Flexible Automation and Intelligent Manufacturing, FAIM 2022, Detroit, MI Ed. Kyoung-Yun Kim, Leslie Monplaisir, Jeremy Rickli, Springer Science and Business Media Deutschland GmbH, 2023. 332-341.
BibTeX: Download