Automatic UATV coordination
TALON’s UATV coordination pilot defines the appropriate methodology for the evaluation of the respective outcomes. In more detail, it will deploy TALON’s E2C AI-Orchestrator in order to surpass the limitations of both centralised and decentralised orchestrators by automatically synthesizing an efficient plan for trajectory planning in a continuous space with a time-varying neighbours and generating safe, dynamically-coupled policies. Specifically, this pilot will i) select the appropriate AI models, datasets and algorithms for the specific problem from TALON’s repository, ii) dynamically formulate and solve the appropriate optimization problem with regard to maximizing energy and data efficiency, and iii) coordinate the UATV operation. Last but not least, it will develop a DT that not only tracks and visualises the execution point of AI models but also evolves based on manual and real-time data.
I5.0 automation & planning
TALON’s I5.0 automation and planning pilot will include the setup, installation and appropriate configuration of a continuous integration (CI)/continuous deployment (CD) environment to ease the development workflows. A set of collaborative workflow tools to facilitate the integration process will be set up (issue tracking, artifacts repository, software building, testing and verification, etc.). This pilot will use as input the software components created within WP3, WP4 and WP4 in order to deploy the demonstration scenarios. The use of automation and configuration management tools will be considered (e.g., Ansible, Terraform), as well as advanced containerization frameworks (e.g., Docker Swarm, Kubernetes) for the deployment of the TALON services. Procedures and practices that assist to the automation of the data management workflows (collection, storage, monitoring, processing and analysis of data) will be also considered in this pilot. Furthermore, the definition of an evaluation methodology for the I5.0 automation and planning demonstrator will be included.
AR/VR for training and maintenance
TALON’s AR/VR for training and maintenance pilot includes the design and definition of an evaluation methodology for this pilot, which covers all preparatory steps that are necessary to set-up this pilot, and coordinate the demonstration, and the evaluation of the pilot. The integration process will follow outputs from technical WPs (WP3 and WP4) and integrate them using state-of-art CI/CD principles while the agility of containers will be utilised. Components of the Security and AI Cognitive layers will also be deployed in in the premises of FACTOR metallurgy plant fulfilling the role of AI orchestrations for the two scenarios and orchestrate AI workloads accelerating visual-aided maintenance and training tasks.
TALON’s human-robot collaboration pilot covers all preparatory steps, which are necessary to set-up this pilot, and coordinate the demonstration, and the evaluation of the pilot. Particularly, in this pilot, the AI-based models and analytics, concerned with analysis and processing of the available data towards optimising the production orchestration, will be defined. Additionally, this pilot will develop the mechanisms for analysing feedback data from the robots, visualising and projecting them in real-time to the operators using AR/VR technologies. Finally, this pilot will analyse the processes and equipment associated with the production and map them to corresponding DT models. The leverage of the DT models, along with AI-based evaluation methodologies, will enable the testing and validation of new and/or improved production processes, prior to their practical implementation.
This project has received funding from the European Union’s Horizon Europe Research and Innovation programme under grant agreement No. 101070181.