Skip to main content
Ir a la página de inicio de la Comisión Europea (se abrirá en una nueva ventana)
español español
CORDIS - Resultados de investigaciones de la UE
CORDIS

Cooperative robotics powered by AI and data for flexible production cells

Periodic Reporting for period 1 - COROB (Cooperative robotics powered by AI and data for flexible production cells)

Período documentado: 2023-10-01 hasta 2025-03-31

The COROB project is focused on developing a flexible, cooperative, and intelligent multirobotic system for arc welding-based manufacturing processes, specifically targeting both joining and additive manufacturing applications. The objective is to introduce new operational capabilities that enhance efficiency and flexibility in industrial production, aligning with the evolving demands of Industry 5.0.
Traditional welding methods often require custom jigs and fixtures for each part, which is especially challenging and costly in high-mix, low-volume (HMLV) production settings. The COROB project addresses this by implementing a jigless robotic welding approach. In this setup, two manipulator robots are used to precisely position and align the workpieces, while a third robot performs the welding operation. This eliminates the need for dedicated physical jigs, enabling rapid reconfiguration for different parts and geometries, and significantly increasing production flexibility. This approach reduces tooling costs, shortens lead times, and is particularly advantageous for custom or small-batch manufacturing scenarios.
The project also leverages Wire Arc Additive Manufacturing (WAAM) for the repair of highvalue components. WAAM is a directed energy deposition process that uses an electric arc to melt and deposit metal wire layer by layer. The COROB system features a processing robot equipped with the WAAM torch, complemented by a rotational unit for workpiece holding and an auxiliary collaborative robot for handling auxiliary manufacturing tasks (e.g. 3D scanning, preheating, others). This integrated multi-robot cell is capable of addressing the repair of existing high-value products, with a focus on tooling. In summary, the COROB project will demonstrate the transformative potential of cooperative multi-robot systems for both jigless welding and WAAM-based repair of high-value manufactured parts.

The technical objectives include:
-TO1. Development of a cooperative multi-robot system for flexible manufacturing (ER1)
-TO2. Development of advanced control system (ER2)
-TO3. Integration of methods for quality inspection (ER3)
-TO4. Development of Process Inspection systems for complex environments (ER4)
-TO5. Development of trainable and reconfigurable AI models (ER5)
-TO6. Digital platform (ER6)
-TO7. Experimental validation of the solution in two semi-industrial use cases (TRL6) determining KPIs related to productivity, quality and sustainability: multi-robot cooperative welding and multi-robot cell for tooling repair using WAAM

The results of the project include (i) 17 SMEs/start-ups funded to develop innovative digital solutions, (ii) development of business plans to their most promising solutions, and (iii) development of a cooperative multi-robot system for flexible manufacturing integrating digital solutions developed by both consortium partners and external parties tested in 2 pilots.
The main activities for each active WP during this period are pinpointed next:
• WP1: successful development of two independent robotic cells designed for flexible arc welding-based manufacturing. Each cell has been configured to perform either joining or additive manufacturing tasks, supporting cooperative multi-robot operations. Both cells are fully operational and have been validated in real conditions, providing a functional foundation for future equipment integration
• WP2: the different possible defects for the welding and the WAAM case have been identified along with the different process variables that create/correct these defects, process input space. The sensing requirements have been defined, the sensors have been selected and a simple visualization dashboard has been developed. A digital tool for ML QA model training/deployment, has been developed integrated with MLOps Level 2 capabilities, focusing on non-AI expert users. Finally, preliminary work has been done on the identification of the possible AL strategies and, some preliminary RL control algorithms have been developed and tested.
• WP3: an architecture has been defined for the overall data management with dedicated implementation per pilot case. Moreover, data connectors utilizing most prominent industrial transfer protocols have been implemented along with the refactoring of the edge part that led to a new platform version.
• WP4:an integration plan for the internal and external developments has been defined to help with coordinating and define cross-correlations among the different developments.
• WP6: the successful launch of both Open Calls, the selection of the participants, and the ongoing mentoring and follow-up of the development by the Third Parties currently in progress as planned. The external solutions are currently being implemented by the Third Parties
• WP7: a robust communication and dissemination strategy has been established, including launching a website, active social media presence, regular newsletters, and participation in major events. Community-building efforts led to a thriving Discord group and a Supportive Partners program, while targeted campaigns boosted engagement among engineering professionals. Additionally, the team focused on developing value propositions and business models to prepare COROB’s solutions for market entry.
• WP8: project coordination and management activities have been developed as planned and no deviations have been reported.

In addition, milestones 1, 2 and 3 have been accomplished during the period:
• D&E&C plan defined.
• COROB Community launched.
• Multi-robot cells operative for arc-based welding processes

The main achievements in the M1-M18 period, are listed next:
• Two fully functional multi-robotic cells have been developed.
• Developed a tool based on function block programming for efficient robot allocation.
• Held sessions on human-centric approach and human interface interactions.
• Defined quality indicators and process input space for both pilots.
• A Machine Learning Operations (MLOps) tool for quality assessment has been deployed.
• The data architecture for the data management platform has been set and the connectors have been developed.
• Successfully launched two open calls and conducted an exhaustive selection process for selecting 17 finalists for the Support Programme, currently on going.
• Established a comprehensive dissemination and communication strategy, including the COROB community.
Consortium members duing the last General Assembly meeting in Valladolid
Mi folleto 0 0