Deep Reinforcement Learning for Robot Batching Optimization and Flow Control

Max Hildebrand, Rasmus Skovgaard Andersen, Simon Bøgh

Publikation: Bidrag til tidsskriftKonferenceartikel i tidsskriftForskningpeer review

9 Citationer (Scopus)
55 Downloads (Pure)

Abstract

Robot batching is an optimization problem found in many industrial applications. Current state-of-the-art approaches utilize a combination of heuristic based parameters and statistical analysis. This approach necessitates many tunable parameters, which again provides challenges when delivering systems to new customers. We challenge current state-of-the-art in statistical approaches by presenting a novel application of a policy gradient method for a Deep Reinforcement Learning (DRL/RL) agent. We have developed a Unity simulation framework of an existing robot- batching cell, on which a RL agent is able to successfully train and obtain a policy for performing robot batching, using a tabula rasa approach. The trained agent is capable of packaging 47.86% of 1218 total batches within the prescribed tolerances, with a positive give-away of 8.76%. The application of DRL in performing robot batching is to the authors knowledge the first of its kind.
OriginalsprogEngelsk
TidsskriftProcedia Manufacturing
Vol/bind51
Sider (fra-til)1462-1468
Antal sider7
ISSN2351-9789
DOI
StatusUdgivet - nov. 2020
Begivenhed30th International Conference on Flexible Automation and Intelligent Manufacturing - Athens, Grækenland
Varighed: 15 jun. 202118 jun. 2021
https://www.faimconference.org/

Konference

Konference30th International Conference on Flexible Automation and Intelligent Manufacturing
Land/OmrådeGrækenland
ByAthens
Periode15/06/202118/06/2021
Internetadresse

Fingeraftryk

Dyk ned i forskningsemnerne om 'Deep Reinforcement Learning for Robot Batching Optimization and Flow Control'. Sammen danner de et unikt fingeraftryk.

Citationsformater