Pseudo-Labeling and Contextual Curriculum Learning for Online Grasp Learning in Robotic Bin Picking

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Le, Huy, Schillinger, Philipp, Gabriel, Miroslav, Qualmann, Alexander, Vien, Ngo Anh
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914702100004864
author Le, Huy
Schillinger, Philipp
Gabriel, Miroslav
Qualmann, Alexander
Vien, Ngo Anh
author_facet Le, Huy
Schillinger, Philipp
Gabriel, Miroslav
Qualmann, Alexander
Vien, Ngo Anh
contents The prevailing grasp prediction methods predominantly rely on offline learning, overlooking the dynamic grasp learning that occurs during real-time adaptation to novel picking scenarios. These scenarios may involve previously unseen objects, variations in camera perspectives, and bin configurations, among other factors. In this paper, we introduce a novel approach, SSL-ConvSAC, that combines semi-supervised learning and reinforcement learning for online grasp learning. By treating pixels with reward feedback as labeled data and others as unlabeled, it efficiently exploits unlabeled data to enhance learning. In addition, we address the imbalance between labeled and unlabeled data by proposing a contextual curriculum-based method. We ablate the proposed approach on real-world evaluation data and demonstrate promise for improving online grasp learning on bin picking tasks using a physical 7-DoF Franka Emika robot arm with a suction gripper. Video: https://youtu.be/OAro5pg8I9U
format Preprint
id arxiv_https___arxiv_org_abs_2403_02495
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Pseudo-Labeling and Contextual Curriculum Learning for Online Grasp Learning in Robotic Bin Picking
Le, Huy
Schillinger, Philipp
Gabriel, Miroslav
Qualmann, Alexander
Vien, Ngo Anh
Robotics
Artificial Intelligence
The prevailing grasp prediction methods predominantly rely on offline learning, overlooking the dynamic grasp learning that occurs during real-time adaptation to novel picking scenarios. These scenarios may involve previously unseen objects, variations in camera perspectives, and bin configurations, among other factors. In this paper, we introduce a novel approach, SSL-ConvSAC, that combines semi-supervised learning and reinforcement learning for online grasp learning. By treating pixels with reward feedback as labeled data and others as unlabeled, it efficiently exploits unlabeled data to enhance learning. In addition, we address the imbalance between labeled and unlabeled data by proposing a contextual curriculum-based method. We ablate the proposed approach on real-world evaluation data and demonstrate promise for improving online grasp learning on bin picking tasks using a physical 7-DoF Franka Emika robot arm with a suction gripper. Video: https://youtu.be/OAro5pg8I9U
title Pseudo-Labeling and Contextual Curriculum Learning for Online Grasp Learning in Robotic Bin Picking
topic Robotics
Artificial Intelligence
url https://arxiv.org/abs/2403.02495