A Machine Learning Approach for Collaborative Robot Smart Manufacturing Inspection for Quality Control Systems

被引：63

作者：

Brito, Thadeu ^{[1
]}

Queiroz, Jonas ^{[1
]}

Piardi, Luis ^{[1
]}

Fernandes, Lucas A. ^{[2
]}

Lima, Jose ^{[1
,3
]}

Leitao, Paulo ^{[1
]}

机构：

[1] Inst Politecn Braganca, Res Ctr Digitalizat & Intelligent Robot CeDRI, Campus Santa Apolonia, P-5300253 Braganca, Portugal

[2] Fed Univ Technol Parana UTFPR, Curitiba, Parana, Brazil

[3] INESC TEC INESC Technol & Sci, Porto, Portugal

来源：

30TH INTERNATIONAL CONFERENCE ON FLEXIBLE AUTOMATION AND INTELLIGENT MANUFACTURING (FAIM2021) | 2020年 / 51卷

关键词：

Collaborative Robots; Quality Control Systems; Reinforcement Learning; Human-robot Interaction; Actor-Critic; Robot Learning;

D O I：

10.1016/j.promfg.2020.10.003

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

The 4t h industrial revolution promotes the automatic inspection of all products towards a zero-defect and high-quality manufacturing. In this context, collaborative robotics, where humans and machines share the same space, comprises a suitable approach that allows combining the accuracy of a robot and the ability and flexibility of a human. This paper describes an innovative approach that uses a collaborative robot to support the smart inspection and corrective actions for quality control systems in the manufacturing process, complemented by an intelligent system that learns and adapts its behavior according to the inspected parts. This intelligent system that implements the reinforcement learning algorithm makes the approach more robust once it can learn and be adapted to the trajectory. In the preliminary experiments, it was used a UR3 robot equipped with a Force-Torque sensor that was trained to perform a path regarding a product quality inspection task. (C) 2020 The Authors. Published by Elsevier Ltd.

引用

页码：11 / 18

页数：8

共 19 条

[1] An application of collaborative robots in a food production facility [J].

Accorsi, R. ;

Tufano, A. ;

Gallo, A. ;

Galizia, F. G. ;

Cocchi, G. ;

Ronzoni, M. ;

Abbate, A. ;

Manzini, R. .

29TH INTERNATIONAL CONFERENCE ON FLEXIBLE AUTOMATION AND INTELLIGENT MANUFACTURING (FAIM 2019): BEYOND INDUSTRY 4.0: INDUSTRIAL ADVANCES, ENGINEERING EDUCATION AND INTELLIGENT MANUFACTURING, 2019, 38 :341-348

[2] Self-learning Processes in Smart Factories: Deep Reinforcement Learning for Process Control of Robot Brine Injection [J].

Andersen, Rasmus E. ;

Madsen, Steffen ;

Barlo, Alexander B. K. ;

Johansen, Sebastian B. ;

Nor, Morten ;

Andersen, Rasmus S. ;

Bogh, Simon .

[3] Transferring Human Manipulation Knowledge to Industrial Robots Using Reinforcement Learning [J].

Arana-Arexolaleiba, N. ;

Urrestilla-Anguiozar, N. ;

Chrysostomou, D. ;

Bogh, S. .

[4]

Brito T., 2019, THESIS I POLITECNICO

[5]

Lillicrap T.P., 2015, arXiv

[6] Teaching Smart Production: An insight into the Learning Factory for Cyber-Physical Production Systems (LVP) [J].

Merkel, Lukas ;

Atug, Johannes ;

Merhar, Laura ;

Schultz, Cedric ;

Braunreuther, Stefan ;

Reinhart, Gunther .

7TH CONFERENCE ON LEARNING FACTORIES (CLF 2017), 2017, 9 :269-274

[7] Hybrid Robotic Reinforcement Learning for Inspection/Correction Tasks [J].

Nasereddin, Hoda ;

Knapp, Gerald M. .

25TH INTERNATIONAL CONFERENCE ON PRODUCTION RESEARCH MANUFACTURING INNOVATION: CYBER PHYSICAL MANUFACTURING, 2019, 39 :406-413

[8] Reinforcement Learning for Manipulators Without Direct Obstacle Perception in Physically Constrained Environments [J].

Ossenkopf, Marie ;

Ennen, Philipp ;

Vossen, Rene ;

Jeschke, Sabina .

27TH INTERNATIONAL CONFERENCE ON FLEXIBLE AUTOMATION AND INTELLIGENT MANUFACTURING, FAIM2017, 2017, 11 :329-337

[9]

Pane YP, 2016, IEEE DECIS CONTR P, P5819, DOI 10.1109/CDC.2016.7799164

[10] A review of the meanings and the implications of the Industry 4.0 concept [J].

Pereira, A. C. ;

Romero, F. .

MANUFACTURING ENGINEERING SOCIETY INTERNATIONAL CONFERENCE 2017 (MESIC 2017), 2017, 13 :1206-1214

← 1 2 →