CORDIS - EU research results
CORDIS

Fog Computing for Robotics and Industrial Automation

Periodic Reporting for period 2 - FORA (Fog Computing for Robotics and Industrial Automation)

Reporting period: 2019-10-01 to 2021-09-30

We are witnessing a new industrial revolution, Industry 4.0 which brings increased productivity and flexibility, mass customization, reduced time-to-market, improved product quality, innovations and new business models. Although Europe has been undergoing a process of deindustrialization, 80% of its exports come from manufacturing, which is responsible for 33 million jobs. The European Commission has set as a target that 20% of value added should come from manufacturing.

Industry 4.0 requires the convergence of Operational and Information Technologies (OT & IT), which use different computation and communication technologies. Cloud Computing cannot be used for OT involving industrial applications since it cannot guarantee stringent non-functional requirements, e.g. dependability, trustworthiness and timeliness. Instead, a new computing paradigm, called Fog Computing, is envisioned as an architectural means to realize the IT/OT convergence.

Fog Computing for Robotics and Industrial Automation, FORA, was a European Training Network which focused on future industrial automation architectures and applications based on an emerging technology, called Fog Computing, that enables the design of new time-sensitive industrial applications. The project was an interdisciplinary, international, intersectoral network that aimed to train the next generation of researchers in Fog Computing with applicability to industrial automation and manufacturing during the period 2017-2021.

The FORA ETN project trained 15 interdisciplinary scientists to gain knowledge in both the Information Technology and Operation Technology domains and understand the demands of the industry that can have an impact at the European level. FORA’s researchers received integrated training across key areas necessary to fully realize the potential of Fog Computing for Industry 4.0 and moved between academic and industrial environments to promote interdisciplinary and intersectoral learning.

The FORA research objectives focused on: future industrial automation architectures and applications based on Fog Computing, deterministic virtualization and execution, deterministic wired and wireless communication, resource provisioning and resource management, Service-Oriented Architecture (SOA), real-time data analytics and security. The result is an open architecture built on open source and open standards, e.g. Time-Sensitive Networking (TSN), OPC Unified Architecture (UA) and 5G. The research took place at the intersection of several research areas: Cloud Computing and Software Engineering, Industrial Automation and Real-Time Systems, and Machine Learning, Security and Safety.
FORA trained 15 Early-Stage Researchers (ESRs) in Fog Computing technologies to deliver Industrial Automation solutions. FORA has provided scientific and transferable skills training to ESRs based on a strong collaboration between academic and industrial partners. FORA has organized three summer schools and a tutorial at the Embedded System Week conference.

Regarding research and results, the main scientific objective of the FORA project was to develop deterministic Fog Computing technologies to be used in industrial automation and robotics solutions for Industry 4.0. We have developed a reference system architecture for Fog Computing, based on deterministic virtualization and networking, and implement open-source prototype Fog Computing Nodes. We have developed Resource Management mechanisms and middleware for deploying scalable Fog Computing applications, while guaranteeing the non-functional properties of the virtualized industrial control applications. We have proposed approaches for assuring the safety and security of the Fog Computing platform. Finally, we have demonstrated how the platform can be used to develop industrial control applications and data analytics applications.

The results have been published in close to 90 publications in journals, conferences, technical reports, and book chapters. All the scientific publications are open access. We have been involved in over 100 dissemination activities, from presentations at conferences, training activities open to external participants, to guest lectures and participation in outreach activities. For example, we have organized two workshops on Fog Computing and IoT (2019, 2020). We have organized our Final Project Conference as a Special Session at ETFA 2021. The FORA results have been presented in the European Researcher’s Night 2019 (Vienna) and via Forskarfredag, a Swedish equivalent of European Researchers’ Night, held online in September 2021.

The universities have exploited the training materials and the results in their graduate-level teaching and by strengthening their academic profile. TTTech has exploited the results via an improved reference architecture focusing on deterministic virtualization solutions based on hypervisors and dynamic separation kernels, which will enable fog computing solutions to be more flexible and agile to the requirements of the industry. SYSGO has exploited the results in their PikeOS hypervisor product extended with hardware-assisted security techniques for safety and real-time critical devices, via an integration of the implemented techniques as module in the existing real-time hypervisor. ABB has exploited the results to evaluate the benefits of the fogification of the next generation of industrial robotics applications, with the aim of transitioning from single core platforms to more complex architectures in automation applications.
FORA has achieved progress beyond the state-of-the-art via the work in 3 research work-packages (WPs): WP1 Fog Computing Platform; WP2 Resource Management and Middleware; WP3 Dependability Services and Application Modeling. We highlight below the progress per WP. WP1 has developed a Fog Node (FN) reference architecture, allowing the execution of applications with differing requirements, in particular in the domains of real-time response, safety, and security. WP2 developed a distributed fog middleware which provides the means to organize a fog landscape, taking into account the demands of the system. In WP3 we addressed the challenge of establishing secure communication channels between fog nodes and rest of the system. We also provided novel methods for intrusion detection, anomaly detection and control flow monitoring. We developed novel fault tolerant techniques as well as safety monitors and reconfiguration. New methods using active machine learning and semi supervised learning on edge devices and federated learning on the fog nodes were also developed.

FORA has delivered a Fog Computing Platform for Industrial IoT based on open standards and open source, which achieves the IT and OT convergence, enabling novel Industry 4.0 applications and business models. The innovative approach of FORA serves the demand for an interoperable cloud based open digital manufacturing platform to ensure Europeans Industry 4.0 competitiveness and wealth. In addition, excellent market opportunities support FORA’s main innovation of an open Fog Computing platform in multiple markets (e.g. industrial, edge computing, embedded OS, IoT security). FORA’s results set a positive example for European industry supporting the dissemination of FORA results and speeding the transition into Industry 4.0.
Participants in the FORA project
FORA Summer School 2019
FORA Training School (TS1), June 2018
FORA Online Summer School 2020