A Novel Policy Distillation With WPA-Based Knowledge Filtering Algorithm for Efficient Industrial Robot Control

Advanced factories strongly need autonomous control methods of manufacturing robots flexibly responding to various requirements of the operational engineers. Deep reinforcement learning is more promising technology to support the dynamic factory situations than the legacy static robot control techno...

Full description

Saved in:
Bibliographic Details
Main Authors: Gilljong Shin, Seongjin Yun, Won-Tae Kim
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10723308/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850201481642246144
author Gilljong Shin
Seongjin Yun
Won-Tae Kim
author_facet Gilljong Shin
Seongjin Yun
Won-Tae Kim
author_sort Gilljong Shin
collection DOAJ
description Advanced factories strongly need autonomous control methods of manufacturing robots flexibly responding to various requirements of the operational engineers. Deep reinforcement learning is more promising technology to support the dynamic factory situations than the legacy static robot control technologies are. The machine learning model experiences the computing resource limitation of industrial robots working the complex jobs of intelligent operations including machine vision, action planning and human collaborations. Policy distillation is a kind of model compression schemes of deep reinforcement learning by means of teacher-student model which make a pre-trained teacher model transfer its knowledge to structurally simplified student models in order to enhance the computing efficiency. However, it may have some problems of anomalous knowledge and abnormal robot movements in case the teacher model has a local optimal policy. In this paper, we propose a novel policy distillation with win probability added (WPA) based knowledge filtering algorithm for efficient industrial robot control. The proposed mechanism adopts the WPA method in sports analytics to evaluate the knowledge extracted from the teacher model and to filter bad knowledge out. The filtered knowledge is reconstructed with interpolation to train the student model with high-quality data. We perform the well-designed experiments, which show 11% compression enhancement and 5% reduction in execution time and steps required for the tasks, using a robot arm and a UGV as test environments.
format Article
id doaj-art-39a51c16d0a240d4a2c1479fa16a8d40
institution OA Journals
issn 2169-3536
language English
publishDate 2024-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-39a51c16d0a240d4a2c1479fa16a8d402025-08-20T02:12:01ZengIEEEIEEE Access2169-35362024-01-011215451415452510.1109/ACCESS.2024.348397010723308A Novel Policy Distillation With WPA-Based Knowledge Filtering Algorithm for Efficient Industrial Robot ControlGilljong Shin0https://orcid.org/0009-0008-6073-1928Seongjin Yun1https://orcid.org/0000-0003-4082-2052Won-Tae Kim2https://orcid.org/0000-0003-3426-3792Department of Computer Science and Engineering, Future Convergence Engineering Major, Korea University of Technology and Education, Cheonan, South KoreaDepartment of Computer Science and Engineering, Future Convergence Engineering Major, Korea University of Technology and Education, Cheonan, South KoreaDepartment of Computer Science and Engineering, Future Convergence Engineering Major, Korea University of Technology and Education, Cheonan, South KoreaAdvanced factories strongly need autonomous control methods of manufacturing robots flexibly responding to various requirements of the operational engineers. Deep reinforcement learning is more promising technology to support the dynamic factory situations than the legacy static robot control technologies are. The machine learning model experiences the computing resource limitation of industrial robots working the complex jobs of intelligent operations including machine vision, action planning and human collaborations. Policy distillation is a kind of model compression schemes of deep reinforcement learning by means of teacher-student model which make a pre-trained teacher model transfer its knowledge to structurally simplified student models in order to enhance the computing efficiency. However, it may have some problems of anomalous knowledge and abnormal robot movements in case the teacher model has a local optimal policy. In this paper, we propose a novel policy distillation with win probability added (WPA) based knowledge filtering algorithm for efficient industrial robot control. The proposed mechanism adopts the WPA method in sports analytics to evaluate the knowledge extracted from the teacher model and to filter bad knowledge out. The filtered knowledge is reconstructed with interpolation to train the student model with high-quality data. We perform the well-designed experiments, which show 11% compression enhancement and 5% reduction in execution time and steps required for the tasks, using a robot arm and a UGV as test environments.https://ieeexplore.ieee.org/document/10723308/Advanced factoryindustrial robotpolicy distillationreinforcement learningmodel compression
spellingShingle Gilljong Shin
Seongjin Yun
Won-Tae Kim
A Novel Policy Distillation With WPA-Based Knowledge Filtering Algorithm for Efficient Industrial Robot Control
IEEE Access
Advanced factory
industrial robot
policy distillation
reinforcement learning
model compression
title A Novel Policy Distillation With WPA-Based Knowledge Filtering Algorithm for Efficient Industrial Robot Control
title_full A Novel Policy Distillation With WPA-Based Knowledge Filtering Algorithm for Efficient Industrial Robot Control
title_fullStr A Novel Policy Distillation With WPA-Based Knowledge Filtering Algorithm for Efficient Industrial Robot Control
title_full_unstemmed A Novel Policy Distillation With WPA-Based Knowledge Filtering Algorithm for Efficient Industrial Robot Control
title_short A Novel Policy Distillation With WPA-Based Knowledge Filtering Algorithm for Efficient Industrial Robot Control
title_sort novel policy distillation with wpa based knowledge filtering algorithm for efficient industrial robot control
topic Advanced factory
industrial robot
policy distillation
reinforcement learning
model compression
url https://ieeexplore.ieee.org/document/10723308/
work_keys_str_mv AT gilljongshin anovelpolicydistillationwithwpabasedknowledgefilteringalgorithmforefficientindustrialrobotcontrol
AT seongjinyun anovelpolicydistillationwithwpabasedknowledgefilteringalgorithmforefficientindustrialrobotcontrol
AT wontaekim anovelpolicydistillationwithwpabasedknowledgefilteringalgorithmforefficientindustrialrobotcontrol
AT gilljongshin novelpolicydistillationwithwpabasedknowledgefilteringalgorithmforefficientindustrialrobotcontrol
AT seongjinyun novelpolicydistillationwithwpabasedknowledgefilteringalgorithmforefficientindustrialrobotcontrol
AT wontaekim novelpolicydistillationwithwpabasedknowledgefilteringalgorithmforefficientindustrialrobotcontrol