Abstract

The project is concerned with research in adaptive learning algorithms. The algorithms will enable robots to react to variations in input, perform ongoing corrections during operations and make adjustments to subsequent operations based on evaluation of the output.
In processes such as the slaughtering of animals and processing of meat, the carcasses and cuts vary in size and composition. The traditional control systems for automatic processes in slaughterhouses rely on strict standardization of these products. When this cannot be guaranteed, automation is either abandoned or large margins of error are introduced, resulting in high cost and suboptimal utilization of the resources. Moreover, customers require that the end-product is customized according to their individual preferences, something traditional automated systems have difficulties handling. With large variations in both input and output, an adaptive robot platform is preferable to more dedicated production machinery.
Original languageEnglish
Title of host publication6th Aalborg U Robotics Workshop
Number of pages1
Publication date27 Nov 2017
Pages16
Publication statusPublished - 27 Nov 2017
Event6th Aalborg U Robotics Workshop - Aalborg, Denmark
Duration: 27 Nov 201727 Nov 2017

Conference

Conference6th Aalborg U Robotics Workshop
CountryDenmark
CityAalborg
Period27/11/201727/11/2017

Fingerprint

Robot learning
Intelligent robots
Meats
Adaptive algorithms
Standardization
Learning algorithms
Machinery
Animals
Automation
Robots
Control systems
Processing
Chemical analysis
Costs

Cite this

@inbook{42e57e678d134a86a8a0a2e0ffa1da06,
title = "Adaptive and Self-learning Slaughter Robots",
abstract = "The project is concerned with research in adaptive learning algorithms. The algorithms will enable robots to react to variations in input, perform ongoing corrections during operations and make adjustments to subsequent operations based on evaluation of the output.In processes such as the slaughtering of animals and processing of meat, the carcasses and cuts vary in size and composition. The traditional control systems for automatic processes in slaughterhouses rely on strict standardization of these products. When this cannot be guaranteed, automation is either abandoned or large margins of error are introduced, resulting in high cost and suboptimal utilization of the resources. Moreover, customers require that the end-product is customized according to their individual preferences, something traditional automated systems have difficulties handling. With large variations in both input and output, an adaptive robot platform is preferable to more dedicated production machinery.",
author = "Philipsen, {Mark Philip} and Andersen, {Rasmus Skovgaard} and Ole Madsen and Moeslund, {Thomas B.}",
year = "2017",
month = "11",
day = "27",
language = "English",
pages = "16",
booktitle = "6th Aalborg U Robotics Workshop",

}

Philipsen, MP, Andersen, RS, Madsen, O & Moeslund, TB 2017, Adaptive and Self-learning Slaughter Robots. in 6th Aalborg U Robotics Workshop. pp. 16, 6th Aalborg U Robotics Workshop, Aalborg, Denmark, 27/11/2017.

Adaptive and Self-learning Slaughter Robots. / Philipsen, Mark Philip; Andersen, Rasmus Skovgaard; Madsen, Ole; Moeslund, Thomas B.

6th Aalborg U Robotics Workshop. 2017. p. 16.

Research output: Contribution to book/anthology/report/conference proceedingConference abstract in proceedingResearch

TY - ABST

T1 - Adaptive and Self-learning Slaughter Robots

AU - Philipsen, Mark Philip

AU - Andersen, Rasmus Skovgaard

AU - Madsen, Ole

AU - Moeslund, Thomas B.

PY - 2017/11/27

Y1 - 2017/11/27

N2 - The project is concerned with research in adaptive learning algorithms. The algorithms will enable robots to react to variations in input, perform ongoing corrections during operations and make adjustments to subsequent operations based on evaluation of the output.In processes such as the slaughtering of animals and processing of meat, the carcasses and cuts vary in size and composition. The traditional control systems for automatic processes in slaughterhouses rely on strict standardization of these products. When this cannot be guaranteed, automation is either abandoned or large margins of error are introduced, resulting in high cost and suboptimal utilization of the resources. Moreover, customers require that the end-product is customized according to their individual preferences, something traditional automated systems have difficulties handling. With large variations in both input and output, an adaptive robot platform is preferable to more dedicated production machinery.

AB - The project is concerned with research in adaptive learning algorithms. The algorithms will enable robots to react to variations in input, perform ongoing corrections during operations and make adjustments to subsequent operations based on evaluation of the output.In processes such as the slaughtering of animals and processing of meat, the carcasses and cuts vary in size and composition. The traditional control systems for automatic processes in slaughterhouses rely on strict standardization of these products. When this cannot be guaranteed, automation is either abandoned or large margins of error are introduced, resulting in high cost and suboptimal utilization of the resources. Moreover, customers require that the end-product is customized according to their individual preferences, something traditional automated systems have difficulties handling. With large variations in both input and output, an adaptive robot platform is preferable to more dedicated production machinery.

M3 - Conference abstract in proceeding

SP - 16

BT - 6th Aalborg U Robotics Workshop

ER -