AI application requires a lot of computation power and memory footprint. That is why, it is not new that, to deploying AI at the edge, it is required to have high-end processors, FPGAs or GPUs. Until recently, low-end devices receive a boost up in the performance, which enables them the capability of running an AI application with limited features. This talk will give the audience an overview of the development states, software aspects, hardware aspects that enable them to run an AI application

Industrial communication systems evolve towards standardized Time-Sensitive Networking (TSN). However, the flexibilization of manufacturing processes often require independency of wired connectivity. The Edge Computing paradigm and a TSN over 5G technology stack are interesting approaches to address this challenge.

Great insights and information resides within edge sensors and actuators on the factory floor or process control environment. Connectivity is the key to accessing and actioning these insights. This talk explores the potential connectivity solutions that will bring the intelligent edge to reality. This presentation discusses the transition from existing field bus technologies to the new 10BASE-T1L Ethernet technology (IEEE standard 802.3cg-2019 / 10BASE-T1L) within the process automation environment. It will outline the use of Ethernet over single twisted pair cabling of up to 1km in length while also adhering to the intrinsically safe, Zone 0 requirement of certain applications. This talk will also explore the opportunities presented by enterprise wide connectivity, while outlining the challenges on the horizon in connecting and leveraging edge intelligence. Connectivity will unleash the true power of edge intelligence and this presentation will explore how we make this a reality.

Throughout this paper, we will talk about the what and how that make IoT solutions complex; it’s not as simple as taking a puzzle you already have completed and adding a few additional pieces to expand the puzzle’s picture. The ability to capitalize on the already-established edge pieces, and then customize and integrate, will make the difference.

It is presented how the application of ML in industry can be democratized: Domain experts are enabled to create ML solutions without expert knowledge in ML. This is possible by combining AutoML with the process and machine knowledge of domain experts.

Today’s IoT solutions are device centric, they are very limited to leverage context. In this session, I am going to show how digital twins of the reality can be built with the Open Sourced Digital Twin Definition Language, contextualized and fundamentally simplify IoT architectures and applications through a live execution environment.

Classical automation concepts with the PLC as the center often have low scalability and high programming complexity. If one thinks here in terms of decentralized reusable modules that bring their necessary logic directly into the devices, exchange process data with each other via the EDGE and are managed via the cloud, one obtains a scalable cost and space-saving solution that no longer requires a central PLC.

AIfES is a machine learning framework where the algorithms are optimized for resource-limited hardware such as microcontrollers. The integration of problem-based prior knowledge and hierarchical structures allows small and efficient implementations.

Todays camera enabled multimedia SoCs are getting more and more complicated and embedded solution architects and system developers have to deal with various challenges on the hardware side, and very extensive software frameworks all the way up from device drivers to networking stacks and full AI stacks on the other side. This speech will give a good overview how such a design might look like and tap in each of the needed building blocks, mainly from a software and system architectural point of view.

AVI has implemented a Deep-Learning hard- and software toolbox to realize deep-learning on the edge applications based on deep-learning accelerator software tools and a unique chip architecture to process data with high speed and very low power consumption.

The presentation covers the following aspects: consideration of different Internet access technologies and wireless technologies for condition monitoring solutions: When is WLAN, Lora, Sigfox, or mobile radio best suited; considered depending on the application; process monitoring and automation with sensor solutions; challenges posed by radio; application example: motor monitoring at Murrelektronik.

Based on a practical example viable for small companies I want to show a reproducible data science workflow from the design of a laboratory experiment to a machine-learning model on the edge device using freely available tools - such as jupyter notebooks with keras or the X-CUBE-AI library.

For to integrate AI as a critical part in solutions with respect to functional safety, different scopes are currently worked on. Because of its importance in development of systems with respect to functional safety, one scope is about the “development process and AI. This seems also to be the more simple part, because in this field there is a lot of experience on how to design “proper” processes to develop functional safe systems. More difficult it is to define techniques and measures to follow when AI is to be designed as a critical part with respect to functional safety. Fundamental research is needed to give the needed confidence in such solutions integrating AI in functional safe systems. The presentation about ongoing activities in the areas “process design” as well as “techniques and measures” should result in starting a discussion about if these activities will result in more confidence in solutions with AI as a critical part with respect to functional safety.