Driving Industry 4.0 at Distributed Edges with Cloud Orchestration
AMQP and MQTT that are implemented in several open source projects from the Eclipse Foundation ( Paho , Mosquitto ), Apache ( ActiveMQ ) and , others like RabbitMQ and HiveMQ . Or OPC- UA implementations within the open source projects of OPC-UA Foundation [ 26 ] [ 27 ]. Eclipse IoT WG [ 28 ] drives many of the standard implementations as open source . Together with Docker containers , open source components such as Kubernetes and Helm are setting industrial standard for building the infrastructure layers of modern edge and cloud computing . Open source projects such as K3s or KubeEdge are adapting Kubernetes to the specific requirements of edge computing implementations . Other examples of open source initiatives driving de-facto standards for edge computing are : Eclipse Foundation ’ s Edge Native WG [ 29 ] [ 30 ], Linux Foundation ’ s LF Edge [ 31 ], and OpenInfra Foundation [ 32 ].
The Asset-Administration Shell ( AAS ) is a prominent industry standard defining metamodels and semantics ( e . g ., asset ID , attributes ) to describe the asset it is representing ( e . g ., machines , AGVs , robots ) and , hence , to realize the digital twin 1 in context of Industry 4.0 [ 36 ]. While it has initially been specified by the Plattform Industrie 4.0 [ 37 ] [ 38 ], the reference implementations are driven by the Industrial Digital Twin Association ( IDTA ) [ 39 ] [ 40 ] using open source as the development model [ 25 ]. The AAS consists of a set of submodels that allow the representation of specific aspects related to asset and its use case ( e . g . technical data , operational data ) [ 36 ] [ 41 ] [ 37 ] [ 38 ]. In general , AAS addresses the rather heterogenous edge computing landscape ( devices , software , industry protocols ).
Assets are then consistently described vertically in all layers ( asset-edge-cloud ) and horizontally in all phases from D2O . In this context , AAS facilitates and ensures the cross-company interoperability of on-boarding scenarios or the application , deployment , and orchestration of digital twin at the edge layer . Although AAS already provides many opportunities in terms of new scenarios and technologies for Industry 4.0 solutions , the specifications of these concepts are still evolving .
Thus , solutions exploring AAS at the edge to create autonomous , distributed systems still need to be discussed in future works . Particularly , how AAS supports backend data integration from different data sources ( cloud , other edge nodes and assets distributed within various systems on the shop floor ) using different standards ( e . g ., OPC-UA , AutomationML ) at the edge .
In context of innovative data-driven business models for Industry 4.0 , International Data Spaces ( IDS ) [ 42 ] [ 43 ] are setting the standards that allow a sovereign and secure data exchange between trusted partners in a federated approach [ 18 ]. At the plant ( edge layer and IoT assets ), IDS enables that data providers ( owners ) have a higher control over the data and its usage within industrial data spaces ( e . g ., Industry 4.0 data space [ 44 ]). IDS is being specified within the activities of
1
Digital twin is defined as a digital representation of an entity , including its attributes and behaviors . The digital twin features bidirectional communication and interaction capabilities to its physical counterpart and environment , preferably in real-time , to ensure its as-manufactured or as-maintained representation . Further details on digital twins can be found in [ 51 ].
- 32 - June 2021