elib
DLR-Header
DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Accessibility | Contact | Deutsch
Fontsize: [-] Text [+]

RODEOS — Robotic Data Ecosystem Semantic Model: Bridging Siloed Robot Systems

Stäbler, Maximilian and Sohlbach, Lukas and Felix, Weidinger and Turnbull, Steffen and Marx-Gómez, Jorge and Langdon, Christoph and Köster, Frank (2025) RODEOS — Robotic Data Ecosystem Semantic Model: Bridging Siloed Robot Systems. In: 24th International Semantic Web Conference, Nara, Japan, November 2–6, 2025, Proceedings. ISWC 2025, 2025-11-02, Nara, Japan. doi: 10.1007/978-3-032-09530-5. ISSN 0302-9743.

[img] PDF
297kB

Abstract

Industry robotics across automotive paint shops, pharmaceutical clean-rooms, and e-commerce warehouses still relies on bespoke data mappings: every new robot arrives with proprietary file formats, capability vocabularies, and safety descriptors, forcing engineering teams into multi-week manual integration cycles. To address this cost and agility gap, we introduce RODEOS—RObotic Data EcOsystem Semantic Model, an emerging, vendor-neutral semantic blueprint co-defined by a consortium of 24 industrial partners. RODEOS extends the W3C DCAT-3 core with robotics-specific classes for raw data, model assets, and executable services, while preserving the lightweight authoring demands voiced in interviews with 17 experts drawn from research institutes, industrial end-users, robotics and automation suppliers, system integrators, IT-infrastructure providers, and the machinery-industry association—thereby capturing a truly holistic cross-domain requirements profile. We conduct a qualitative ablation study comparing JSON schemas generated by an LLM with and without RODEOS schema constraints, finding that schema-guided generation improves precision, coverage, and consistency of the output. These contributions—(i) the RODEOS semantic model, (ii) an LLM-assisted authoring workflow, and (iii) initial empirical validation metrics—aim to accelerate robot-cell integration and lay the groundwork for a community-wide semantic robotics ecosystem.

Item URL in elib:https://elib.dlr.de/215878/
Document Type:Conference or Workshop Item (Speech)
Title:RODEOS — Robotic Data Ecosystem Semantic Model: Bridging Siloed Robot Systems
Authors:
AuthorsInstitution or Email of AuthorsAuthor's ORCID iDORCID Put Code
Stäbler, MaximilianMaximilian.Staebler (at) dlr.dehttps://orcid.org/0000-0003-1311-3568218076667
Sohlbach, Lukaslukas.sohlbach (at) vdma.orgUNSPECIFIEDUNSPECIFIED
Felix, Weidingerfelix.weidinger (at) vdma.orgUNSPECIFIEDUNSPECIFIED
Turnbull, Steffensteffen.turnbull (at) dlr.deUNSPECIFIEDUNSPECIFIED
Marx-Gómez, JorgeCarl von Ossietzky Universität OldenburgUNSPECIFIEDUNSPECIFIED
Langdon, ChristophDrucker School of Business, U.S.A.UNSPECIFIEDUNSPECIFIED
Köster, FrankFrank.Koester (at) dlr.dehttps://orcid.org/0000-0002-5422-3109UNSPECIFIED
Date:2 November 2025
Journal or Publication Title:24th International Semantic Web Conference, Nara, Japan, November 2–6, 2025, Proceedings
Refereed publication:Yes
Open Access:Yes
Gold Open Access:No
In SCOPUS:No
In ISI Web of Science:No
DOI:10.1007/978-3-032-09530-5
ISSN:0302-9743
Status:Published
Keywords:industrial robotics, semantic interoperability, LLM-assisted semantic engineering, vendor-neutral semantics
Event Title:ISWC 2025
Event Location:Nara, Japan
Event Type:international Conference
Event Date:2 November 2025
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Transport
HGF - Program Themes:Transport System
DLR - Research area:Transport
DLR - Program:V VS - Verkehrssystem
DLR - Research theme (Project):V - DiVe - Digital organisiertes Verkehrssystem
Location: Ulm
Institutes and Institutions:Institute for AI Safety and Security
Deposited By: Stäbler, Maximilian
Deposited On:19 Aug 2025 13:37
Last Modified:09 Nov 2026 14:22

Repository Staff Only: item control page

Browse
Search
Help & Contact
Information
OpenAIRE Validator logo electronic library is running on EPrints
Website and database design: Copyright © German Aerospace Center (DLR). All rights reserved.