Periodic Reporting for period 2 - ACROBA (AI-Driven Cognitive Robotic Platform for Agile Production environments)
Periodo di rendicontazione: 2022-01-01 al 2022-12-31
As modern industrial robotic systems become smarter and more flexible, using them in small-scale industrial environments is increasingly becoming a viable option. Generally, industrial robotics is tailored towards specific, large-scale operations, making its implementation too complex and costly for smaller operators. The EU-funded ACROBA project aims to develop a groundbreaking concept of scalable cognitive robotics platforms to revolutionise industrial production. It will achieve this through the use of cutting-edge technology and the application of agile manufacturing principles. This will enable the rapid reconfiguration of robotics production platforms, allowing the implementation of advanced, cost-effective robotics solutions in small- and large-scale industrial applications.
ACROBA aims to develop and demonstrate a novel concept of cognitive robotic platforms based on a modular approach able to be smoothly adapted to virtually any industrial scenario applying agile manufacturing principles. The novel industrial platform will be based on the concept of plug-and-produce, featuring a modular and scalable architecture allowing the connection of robotic systems with enhanced cognitive capabilities to deal with cyber physical systems in fast-changing production environments. The platform will take advantage of artificial intelligence and cognitive modules to meet personalisation requirements and enhance mass product customisation through advanced robotic systems capable of self-adapting to the different production needs. A novel ecosystem will be built as a result of this project, enabling the fast and economic deployment of advanced robotic solutions in agile manufacturing industrial lines, especially industrial SMEs.
The characteristics of the ACROBA platform will allow its cost-effective integration and smooth adoption by diverse industrial scenarios to realise their true industrialisation within agile production environments becoming a cost-effective solution for a wide range of Industrial sectors, both inside the consortium and additional industrial sectors that will be addressed in the future. The Project approach will be demonstrated in five industrial large-scale real pilots and tested through twelve dedicated Hackatons and two ACROBA On-Site Labs for technology transfer experiments.
ACROBA fulfils three out of seventeen Sustainable Development Goals (SDGs) contributing to the improvement of society; SDG 9.2. “Promote inclusive and sustainable industrialization and, by 2030, significantly raise industry’s share of employment and gross domestic product” and SDG 10.2. "By 2030, empower and promote the social, economic and political inclusion of all, irrespective of age, sex, disability, race, ethnicity, origin, religion or economic or other status".
ACROBA aims to develop and demonstrate a novel concept of cognitive robotic platforms based on a modular approach able to be smoothly adapted to virtually any industrial scenario applying agile manufacturing principles. The novel industrial platform will be based on the concept of plug-and-produce, featuring a modular and scalable architecture allowing the connection of robotic systems with enhanced cognitive capabilities to deal with cyber physical systems in fast-changing production environments. The platform will take advantage of artificial intelligence and cognitive modules to meet personalisation requirements and enhance mass product customisation through advanced robotic systems capable of self-adapting to the different production needs. A novel ecosystem will be built as a result of this project, enabling the fast and economic deployment of advanced robotic solutions in agile manufacturing industrial lines, especially industrial SMEs.
The characteristics of the ACROBA platform will allow its cost-effective integration and smooth adoption by diverse industrial scenarios to realise their true industrialisation within agile production environments becoming a cost-effective solution for a wide range of Industrial sectors, both inside the consortium and additional industrial sectors that will be addressed in the future. The Project approach will be demonstrated in five industrial large-scale real pilots and tested through twelve dedicated Hackatons and two ACROBA On-Site Labs for technology transfer experiments.
ACROBA fulfils three out of seventeen Sustainable Development Goals (SDGs) contributing to the improvement of society; SDG 9.2. “Promote inclusive and sustainable industrialization and, by 2030, significantly raise industry’s share of employment and gross domestic product” and SDG 10.2. "By 2030, empower and promote the social, economic and political inclusion of all, irrespective of age, sex, disability, race, ethnicity, origin, religion or economic or other status".
The second reporting period (M13-M24) started after the completion of WP1 in M12 that resulted in the ACROBA platform architecture and implementation. During this period, along with WP2-WP6, the platform was successfully deployed in the scenarios and partially integrated with existing equipment. WP2 has focused on the development of the different robot modules given the initial architecture defined in WP1. Perception and control algorithms were developed and integrated into the platform as skills and primitives. Automatic task generation and robot programming based on dummy tools have also been addressed. A graphical user interface was proposed to allow operators to easily customize or create new tasks. For safe collaborative tasks, a toolkit was developed to address the human-robot interaction in a shared workspace. The virtual gym, was created to replicate the different manufacturing tasks. In parallel, a DRL was built to optimize perception and control algorithms. WP3 has been focused on the definition of the agile manufacturing cell in collaboration with WP1 and WP2, and in the implementation of the ACROBA platform for the preparation of some generic tests. The agile robotic cells were developed and demonstrated. For the integration of the modules, a common framework has been implemented and interfaces between every module have been defined and validated. Regarding the Human Factor, a safety concept design has been outlined and effort has been made in order to ensure the fulfilling of the safety requirements. WP4 and WP5 have been moving forward with the definition of the use cases and deployment tasks have begun. WP6 has worked on assessing and exploring the scalability and flexibility of the ACROBA platform. In addition, work on the development of the single-entry point to allow all third parties to adopt ACROBA platform has been evolving. WP7 has focused on developing the project’s online visibility through the ACROBA website and social media and on enhancing the dissemination of the project’s results. The first hackathon was organised successfully. WP8 has worked on the definition of ACROBA’s exploitation strategy. AOSL discussions have started. WP9 has performed the project coordination from the administrative, financial, and technical management standpoint. All the procedures defined and established during the first reporting period have been followed successfully.
ACROBA aims to progress beyond the state of the artin in the following points:
1) ACROBA aims to evolve the current software architectures for advanced robotic systems by adding different functionalities of vital importance in the development of robotic integrated systems, i) innovative definition of the semantic, ii) direct integration of the ACROBA platform with virtual simulation environments (Virtual Gyms) and iii) smooth transition from training environment to real environment
2) ACROBA will allow robots to be embedded neatly in the existing workflows of SMEs, improving robots autonomy (execution of sensor-based robotic skills), its programming (use of dummy tools & automatic generation of paths from CAD files) and its productivity and adaptability (new cognitive capabilities for the planning module)
3) ACROBA will ensure quick autonomous and collaborative cognitive robotic capabilities through the adoption and combination of complementary AI approaches, e.g. Deep Learning , Deep Reinformed Learning, Deep Imitation Learning and Evolutionary Computing.
During the first year, the framework to achieve this has been defined.
ACROBA expected impacts are as follows:
1) Demonstration of the potential for robotics to impact at scale in Agile production
2) Reduction of technical and commercial risk in the deployment of services based on robotic actors within Agile production
3) Greater understanding from the application stakeholders of the potential for deploying robotics
4) Demonstration of platforms operating over extended time periods in near realistic environments and promotion of their use
5) Develop the eco-system around the Agile Production to stimulate deployment
6) Contribution to the development of open, industry-led or de facto standards
During the first year the project has been focused on the development of the architecture and the framework for Acroba. During the next months of the project with the pilot cases running, the measures to maximize the impact will be reviewed and significant contribution is expected. Also, an improvement in communication and dissemination actions will be taken to increase the impact both to reach an even wider public and the stakeholders related to the ACROBA solution.
1) ACROBA aims to evolve the current software architectures for advanced robotic systems by adding different functionalities of vital importance in the development of robotic integrated systems, i) innovative definition of the semantic, ii) direct integration of the ACROBA platform with virtual simulation environments (Virtual Gyms) and iii) smooth transition from training environment to real environment
2) ACROBA will allow robots to be embedded neatly in the existing workflows of SMEs, improving robots autonomy (execution of sensor-based robotic skills), its programming (use of dummy tools & automatic generation of paths from CAD files) and its productivity and adaptability (new cognitive capabilities for the planning module)
3) ACROBA will ensure quick autonomous and collaborative cognitive robotic capabilities through the adoption and combination of complementary AI approaches, e.g. Deep Learning , Deep Reinformed Learning, Deep Imitation Learning and Evolutionary Computing.
During the first year, the framework to achieve this has been defined.
ACROBA expected impacts are as follows:
1) Demonstration of the potential for robotics to impact at scale in Agile production
2) Reduction of technical and commercial risk in the deployment of services based on robotic actors within Agile production
3) Greater understanding from the application stakeholders of the potential for deploying robotics
4) Demonstration of platforms operating over extended time periods in near realistic environments and promotion of their use
5) Develop the eco-system around the Agile Production to stimulate deployment
6) Contribution to the development of open, industry-led or de facto standards
During the first year the project has been focused on the development of the architecture and the framework for Acroba. During the next months of the project with the pilot cases running, the measures to maximize the impact will be reviewed and significant contribution is expected. Also, an improvement in communication and dissemination actions will be taken to increase the impact both to reach an even wider public and the stakeholders related to the ACROBA solution.