kth.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Autonomous agents in Industry 4.0: A self-optimizing approach for automated guided vehicles in Industry 4.0 environments
KTH, School of Electrical Engineering and Computer Science (EECS).
2022 (English)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesis
Abstract [en]

Automated guided vehicles are an integral part of industrial production today. They are moving products to and from shelves in storage warehouses and fetching tools between different workstations in factories. These robots usually follow strict pre-determined paths and are not good at adapting to changes in the environment. Technologies like artificial intelligence and machine learning are currently being implemented in industrial production, a part of what is called Industry 4.0, with the aim of increasing efficiency and automation. Industry 4.0 is also characterized by more connected factory environments, where objects communicate their status, location, and other relevant information to their surroundings. Automated guided vehicles can take advantage of these technologies and can benefit from self-optimizing approaches for better navigation and increased flexibility. Reinforcement learning is used in this project to teach automated guided vehicles to move objects around in an Industry 4.0 warehouse environment. A 10x10 grid world with numerous object destinations, charging stations and agents is created for evaluation purposes. The results show that the agents are able to learn to take efficient routes by balancing the need to finish tasks as fast as possible and recharge their batteries when needed. The agents successfully complete all tasks without running out of battery or colliding with objects in the environment. The result is a demonstration of how reinforcement learning can be applied to automated guided vehicles in Industry 4.0 environments.

Abstract [sv]

Automatiserade styrda fordon är en integrerad del av dagens industriproduktion. De flyttar produkter till och från hyllor i lagerlokaler och hämtar verktyg mellan olika arbetsstationer i fabriker. Dessa robotar följer vanligtvis strikta förutbestämda vägar och är inte bra på att anpassa sig till förändringar i miljön. Teknik som artificiell intelligens och maskininlärning implementeras just nu i industriproduktion, en del av det som kallas Industri 4.0, i syfte om ökad effektivitet och automatisering. Industri 4.0 kännetecknas också av mer uppkopplade fabriksmiljöer, där objekt kommunicerar sin status, plats och annan relevant information till sin omgivning. Automatiserade styrda fordon kan utnyttja de här teknikerna och kan dra nytta av självoptimerande metoder för bättre navigering och ökad flexibilitet. Förstärkningsinlärning används i detta projekt för att lära automatiserade styrda fordon att flytta runt föremål i en Industri 4.0 lagermiljö. En 10x10 stor rut-värld med flertalet destinationer, laddningsstationer och agenter skapas i utvärderingssyfte. Resultaten visar att agenterna kan lära sig att ta effektiva vägar genom att balansera behovet av att slutföra sina uppgifter så fort som möjligt och ladda upp sina batterier när det behövs. Agenterna slutför framgångsrikt sina uppgifter utan att få slut på batteri eller att kollidera med föremål i miljön. Resultatet är en demonstration av hur förstärkningsinlärning kan tillämpas på automatiserade styrda fordon i Industri 4.0-miljöer.

Place, publisher, year, edition, pages
2022. , p. 41
Series
TRITA-EECS-EX ; 2022:288
Keywords [en]
Automated guided vehicles, Industry 4.0, Agents, Reinforcement learning
Keywords [sv]
Automatiserade guidade fordon, Industri 4.0, Agenter, Förstärkningsinlärning
National Category
Computer Engineering
Identifiers
URN: urn:nbn:se:kth:diva-319795OAI: oai:DiVA.org:kth-319795DiVA, id: diva2:1701869
Subject / course
Computer Science
Educational program
Bachelor of Science in Engineering - Computer Engineering
Supervisors
Examiners
Available from: 2022-10-10 Created: 2022-10-07 Last updated: 2022-10-10Bibliographically approved

Open Access in DiVA

fulltext(609 kB)331 downloads
File information
File name FULLTEXT01.pdfFile size 609 kBChecksum SHA-512
e82f3223237f8c59c09a7472164e8f6a0cc9241beb0902a35096bd3f30fa2d8347549883ce5e1a81506bed3a44daf7cb74dab14d7c3e10e4334c511cfb08382a
Type fulltextMimetype application/pdf

By organisation
School of Electrical Engineering and Computer Science (EECS)
Computer Engineering

Search outside of DiVA

GoogleGoogle Scholar
Total: 334 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 309 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf