Automatic Parameter Learning for Easy Instruction of Industrial Collaborative Robots

Lars Carøe Sørensen, Rasmus Skovgaard Andersen, Casper Schou, Dirk Kraft

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

1 Citation (Scopus)

Abstract

The manufacturing industry faces challenges in meeting requirements of flexibility, product variability and small batch sizes. Automation of high mix, low volume productions requires faster (re)configuration of manufacturing equipment. These demands are to some extend accommodated by collaborative robots. Certain actions can still be hard or impossible to manually adjust due to inherent process uncertainties. This paper proposes a generic iteratively learning approach based on Bayesian Optimisation to efficiently search for the optimal set of process parameters. The approach takes into account the process uncertainties by iteratively making a statistical founded choice on the next parameter-set to examine only based on the prior binomial outcomes. Moreover, our function estimator uses Wilson Score to make proper estimates on the success probability and the associated uncertain measure of sparsely sampled regions. The function estimator also generalises the experiment outcomes to the neighbour region through kernel smoothing by integrating Kernel Density Estimation. Our approach is applied to a real industrial task with significant process uncertainties, where sufficiently robust process parameters cannot intuitively be chosen. Using our approach, a collaborative robot automatically finds a reliable solution.
Original languageEnglish
Title of host publicationProceedings - 2018 IEEE International Conference on Industrial Technology, ICIT 2018
Number of pages6
Volume2018-February
PublisherIEEE
Publication dateApr 2018
Pages87-92
ISBN (Electronic)9781509059492
DOIs
Publication statusPublished - Apr 2018
Event2018 IEEE International Conference on Industrial Technology - Lyon, France
Duration: 20 Feb 201822 Feb 2018
http://icit2018.org/en

Conference

Conference2018 IEEE International Conference on Industrial Technology
CountryFrance
CityLyon
Period20/02/201822/02/2018
Internet address

Fingerprint

Robots
Automation
Uncertainty
Industry
Experiments

Keywords

  • Industrial assembly
  • Parameter optimisation

Cite this

Sørensen, L. C., Andersen, R. S., Schou, C., & Kraft, D. (2018). Automatic Parameter Learning for Easy Instruction of Industrial Collaborative Robots. In Proceedings - 2018 IEEE International Conference on Industrial Technology, ICIT 2018 (Vol. 2018-February, pp. 87-92). IEEE. https://doi.org/10.1109/ICIT.2018.8352157
Sørensen, Lars Carøe ; Andersen, Rasmus Skovgaard ; Schou, Casper ; Kraft, Dirk. / Automatic Parameter Learning for Easy Instruction of Industrial Collaborative Robots. Proceedings - 2018 IEEE International Conference on Industrial Technology, ICIT 2018. Vol. 2018-February IEEE, 2018. pp. 87-92
@inproceedings{1f54395d3edf48b79ec57b671bd10f90,
title = "Automatic Parameter Learning for Easy Instruction of Industrial Collaborative Robots",
abstract = "The manufacturing industry faces challenges in meeting requirements of flexibility, product variability and small batch sizes. Automation of high mix, low volume productions requires faster (re)configuration of manufacturing equipment. These demands are to some extend accommodated by collaborative robots. Certain actions can still be hard or impossible to manually adjust due to inherent process uncertainties. This paper proposes a generic iteratively learning approach based on Bayesian Optimisation to efficiently search for the optimal set of process parameters. The approach takes into account the process uncertainties by iteratively making a statistical founded choice on the next parameter-set to examine only based on the prior binomial outcomes. Moreover, our function estimator uses Wilson Score to make proper estimates on the success probability and the associated uncertain measure of sparsely sampled regions. The function estimator also generalises the experiment outcomes to the neighbour region through kernel smoothing by integrating Kernel Density Estimation. Our approach is applied to a real industrial task with significant process uncertainties, where sufficiently robust process parameters cannot intuitively be chosen. Using our approach, a collaborative robot automatically finds a reliable solution.",
keywords = "Industrial assembly, Parameter optimisation",
author = "S{\o}rensen, {Lars Car{\o}e} and Andersen, {Rasmus Skovgaard} and Casper Schou and Dirk Kraft",
year = "2018",
month = "4",
doi = "10.1109/ICIT.2018.8352157",
language = "English",
volume = "2018-February",
pages = "87--92",
booktitle = "Proceedings - 2018 IEEE International Conference on Industrial Technology, ICIT 2018",
publisher = "IEEE",
address = "United States",

}

Sørensen, LC, Andersen, RS, Schou, C & Kraft, D 2018, Automatic Parameter Learning for Easy Instruction of Industrial Collaborative Robots. in Proceedings - 2018 IEEE International Conference on Industrial Technology, ICIT 2018. vol. 2018-February, IEEE, pp. 87-92, 2018 IEEE International Conference on Industrial Technology, Lyon, France, 20/02/2018. https://doi.org/10.1109/ICIT.2018.8352157

Automatic Parameter Learning for Easy Instruction of Industrial Collaborative Robots. / Sørensen, Lars Carøe; Andersen, Rasmus Skovgaard; Schou, Casper; Kraft, Dirk.

Proceedings - 2018 IEEE International Conference on Industrial Technology, ICIT 2018. Vol. 2018-February IEEE, 2018. p. 87-92.

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

TY - GEN

T1 - Automatic Parameter Learning for Easy Instruction of Industrial Collaborative Robots

AU - Sørensen, Lars Carøe

AU - Andersen, Rasmus Skovgaard

AU - Schou, Casper

AU - Kraft, Dirk

PY - 2018/4

Y1 - 2018/4

N2 - The manufacturing industry faces challenges in meeting requirements of flexibility, product variability and small batch sizes. Automation of high mix, low volume productions requires faster (re)configuration of manufacturing equipment. These demands are to some extend accommodated by collaborative robots. Certain actions can still be hard or impossible to manually adjust due to inherent process uncertainties. This paper proposes a generic iteratively learning approach based on Bayesian Optimisation to efficiently search for the optimal set of process parameters. The approach takes into account the process uncertainties by iteratively making a statistical founded choice on the next parameter-set to examine only based on the prior binomial outcomes. Moreover, our function estimator uses Wilson Score to make proper estimates on the success probability and the associated uncertain measure of sparsely sampled regions. The function estimator also generalises the experiment outcomes to the neighbour region through kernel smoothing by integrating Kernel Density Estimation. Our approach is applied to a real industrial task with significant process uncertainties, where sufficiently robust process parameters cannot intuitively be chosen. Using our approach, a collaborative robot automatically finds a reliable solution.

AB - The manufacturing industry faces challenges in meeting requirements of flexibility, product variability and small batch sizes. Automation of high mix, low volume productions requires faster (re)configuration of manufacturing equipment. These demands are to some extend accommodated by collaborative robots. Certain actions can still be hard or impossible to manually adjust due to inherent process uncertainties. This paper proposes a generic iteratively learning approach based on Bayesian Optimisation to efficiently search for the optimal set of process parameters. The approach takes into account the process uncertainties by iteratively making a statistical founded choice on the next parameter-set to examine only based on the prior binomial outcomes. Moreover, our function estimator uses Wilson Score to make proper estimates on the success probability and the associated uncertain measure of sparsely sampled regions. The function estimator also generalises the experiment outcomes to the neighbour region through kernel smoothing by integrating Kernel Density Estimation. Our approach is applied to a real industrial task with significant process uncertainties, where sufficiently robust process parameters cannot intuitively be chosen. Using our approach, a collaborative robot automatically finds a reliable solution.

KW - Industrial assembly

KW - Parameter optimisation

UR - http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8352157&isnumber=8352140

U2 - 10.1109/ICIT.2018.8352157

DO - 10.1109/ICIT.2018.8352157

M3 - Article in proceeding

VL - 2018-February

SP - 87

EP - 92

BT - Proceedings - 2018 IEEE International Conference on Industrial Technology, ICIT 2018

PB - IEEE

ER -

Sørensen LC, Andersen RS, Schou C, Kraft D. Automatic Parameter Learning for Easy Instruction of Industrial Collaborative Robots. In Proceedings - 2018 IEEE International Conference on Industrial Technology, ICIT 2018. Vol. 2018-February. IEEE. 2018. p. 87-92 https://doi.org/10.1109/ICIT.2018.8352157