to Control an IoT-device by a Control Message

The invention relates to a method for automatic creating an asset data model using a pub- lish-subscribe software architecture to control an IoT-device by a control message.

The invention relates further to a platform for automatic creating an asset data model using a publish-subscribe software architecture, and for automatic publishing messages based on the asset data model.

The invention relates further to a system comprising the platform and an IoT-device.

Digitally transforming assets is a priority for manufacturers in the age of Industry 4.0, i.e. Internet-of-Things (IoT). Plant assets are known as for instance technical devices, produc tion equipment or machines, like CNC machines. Computers and automation technology have come together to create game-changing possibilities and the global marketplace has been turned completely on its head. To compete globally, manufacturers are looking to connect factory machines to the cloud and become more flexible in their production to meet customer expectations. However, they're resistant to making changes to their ma chines because doing so seems too risky, expensive, insecure, and intrusive, especially given the wide variety of devices on the plant floor.

Recent technological developments, however, address these barriers to smart factory transformation, helping companies get connected to the cloud to manage data and devic es, gain insights, and leverage machine learning capabilities. The backbone of this connec tivity is the Open Platform Communications Unified Architecture (OPC UA), which sim plifies interoperability and enables manufacturers to get a head start on transforming their assets into smart factories.

OPC UA is a standard that ensures the open connectivity, interoperability, security, and reliability of industrial automation devices and systems. OPC UA is widely recognized as the key communication and data modeling technology for the Industry 4.0 initiative. It works with many software platforms, is completely scalable, and has been widely adopt ed across several industries because of its flexibility.

However, the introduction of OPC UA into smart factories does not lead to solve individ ual process steps, which are often still manually. Today, IoT platforms comprise often data models, which are designed to store known data structures. However, before data being stored, they need to be modelled first. But this procedure is not suitable for OPC UA PubSub data uploading approach. According to OPC UA, PubSub devices, which are sending OPC UA PubSub data first, must define and send meta data and model data.

In software architecture, publish-subscribe (PubSub) is a messaging pattern where send ers of messages, called publishers, do not program the messages to be sent directly to spe cific receivers, called subscribers, but instead categorize published messages into classes without knowledge of which subscribers, if any, there may be. Similarly, subscriber ex press interest in one or more classes and only receive messages that are of interest, with out knowledge of which publishers, if any, there are.

Publish-subscribe is a sibling of the message queue paradigm and is typically one part of a larger message-oriented middleware system. Most messaging systems support both the pub/ sub and message queue models in their API, e.g. Java Message Service (JMS).

This pattern provides greater network scalability and a more dynamic network topology, with a resulting decreased flexibility to modify the publisher and the structure of the pub lished data.

In the publish-subscribe model, subscribers typically receive only a subset of the total messages published. The process of selecting messages for reception and processing is called filtering. There are two common forms of filtering: topic -based and content-based.

In a topic -based system, messages are published to "topics" or named logical channels. Subscribers in a topic-based system will receive all messages published to the topics to which they subscribe, and all subscribers to a topic will receive the same messages. The publisher is responsible for defining the classes of messages to which subscribers can sub scribe.

In a content-based system, messages are only delivered to a subscriber if the attributes or content of those messages matches constraints defined by the subscriber. The subscriber is responsible for classifying the messages.

Some systems support a hybrid of the two; publishers post messages to a topic while sub scribers register content-based subscriptions to one or more topics.

In prior art, IoT platform data modeling needs to be done manually, and before the IoT devices can start to operate, data need to be mapped manually to a data model, which is needed to control the IoT device. However, manual editing of complex data structures needs skilled persons, is time consuming and error-prone. Thus, manual modelling is ex pensive and undesirable.

Thus, it is the objective of the invention to improve further the automatization of smart factories using OPC UA, like the Siemens MindSphere® IoT platform, and to automate manual tasks in prior art in order to eliminate complex manual operations, which can cause errors and prolong the overall production time, in order to reduce cost and produc tion time and further, to increase data integrity and reliability.

The objective of the invention is obtained by a method for automatic creating an asset data model using a publish-subscribe software architecture to control an IoT-device by a con trol message, comprising following steps:

a. receiving an initial message from a communication input means containing meta data and message data, wherein the meta data comprise structure definition data, structure field data and field meta data,

b. creating an aspect type with aspect type data and an asset type with aspect type data from the meta data and creating the asset data model automatically from the aspect type and the asset type by mapping corresponding data, wherein the aspect type data are determined at least from the meta data, structure definition data, structure field data and field meta data, and the asset type data are determined at least from the aspect type data,

c. generating the control message, comprising the asset data model,

d. publishing the control message to the IoT-device using the publish-subscribe communication principle.

An asset is a digital representation of a machine or of an automation system with one or more automation devices connected to the Siemens MindSphere® platform (e.g. PLC). Data collection and data delivery with MindSphere relies on so-called (virtual) assets. An asset can be any element, e.g. as a pump, a motor, a PLC, a complete machine tool, a pro duction line, a robot, a crane, a vehicle, a wind turbine, etc., just to name a few examples. The data describing an asset is collected and transmitted to MindSphere. These data are then made available for further processing and evaluation.

A communication input means in the present context refers to a subsystem, which is pre located before the device carrying out the method according to the invention and which delivers proper data for the subsequent data processing entity. The communication input means for instance can be configured to perform loading data from a storage or to receive by a communication device or simply to capture data by a manual input means using a keyboard and a computer means. According to the invention, the input data are used to generate the initial message, which is done by the communication input means. The initial message comprises the message data, which are subject of the subsequent data processing steps according to the invention.

Aspects are a data modeling mechanism for plant assets. Aspect objects relate to all of data of the plant, and the aspects to specific plant assets. For example, aspects are infor mational items associated with objects, such as I/O definitions, engineering drawings, process graphics, reports, trends, etc. that are assigned to each object in a system. Aspects group related data points based on their logical association. Example: The pump set has an aspect, z. For example, "Energy Consumption", which contains the data points "Pow er," "Power," "Voltage," and so on. The aspect should have a connection to data points and a physical asset. An aspect can comprise several variables.

A system can provide an Aspect server and Connectivity servers on separate nodes or on Application servers, depending on the size of the installation.

A type is a preconfigured template for an asset. Assets assume the properties of the type on which they are based. Within the Type you can define which Aspects are integrated into the template.

Consequently, aspect types and asset types are templates for aspects and assets.

Meta data can comprise structure definition data for defining a data structure of the meta data, or structure field data for defining data fields like data formats, or field meta data for defining a description of the usage of the meta data. Basically, the meta data are in tended to describe the message data of the initial message.

Thus, the objective of the invention is being achieved by an automating step of modeling specific data, which are needed for subsequent manufacturing steps, considering OPC UA PubSub prerequisites. By the automatization in the generation of the model, the efficiency of the programming and the operation at all is improved. Moreover, the data integrity is improved since manual operation in prior art frequently incorporates errors in the used data model. Thus, the method enables to control a system of IoT-devices automatically in a very effi cient and reliable way.

The objective of the invention is obtained also by a platform for automatic creating an asset data model using a publish-subscribe software architecture, and for automatic gen erating and publishing messages based on the asset data model, comprising means con figured to carry out the aforesaid method according to the invention.

Asset data can describe related assets or their functionality and features, like communica tion features. The asset data can comprise an asset data model e.g. to describe dependen cies of asset data of a related asset.

The objective of the invention is obtained also by a system comprising the platform ac cording the preceding claim and one or even a plurality of IoT-devices, wherein the mes senger is configured to control the IoT-device by the published message.

The invention is not restricted to the specific embodiments described in detail herein, but encompasses all variants, combinations and modifications thereof that fall within the framework of the appended claims.

The invention will be explained in more detail with reference to an exemplary embodi ment shown in the accompanying drawings. The figures show in

Fig. 1 an embodiment of the method of the invention,

Fig. 2 an example of created assets and aspect types,

Fig. 3 an example of used meta data,

Fig. 4 an example of a Domain Model of Asset-Aspect Types and Instances,

Fig. 5 an embodiment of Aspect Type for SPSFields,

Fig. 6 a flowchart with an example of automatic creation of Asset- Aspect Types,

Fig. 7 a system arrangement of an embodiment according to the invention.

It is clear, that further not shown parts are necessary for the operation of the platform or system, e.g. means for communication, IO, storage or processing. For the sake of better understanding these parts are not illustrated and described.

The method according to the invention can be executed as a computer program on a com puter platform or system, wherein the platform or system can be a local or a distributed equipment. Moreover, it is clear, that more than one IoT device, i.e. a plurality of IoT devices, can be controlled by the method, the platform and the system according to the invention.

Fig. 1 shows an embodiment of the method according to the invention. The method 100 for automatic creating of an asset data model using a publish-subscribe software architec ture to control an IoT-device by a control message, comprises following steps:

a. receiving 110 an initial message from a communication input means containing meta data and message data, wherein the meta data comprise structure definition data, structure field data and field meta data,

b. creating 120 an aspect type with aspect type data and an asset type with aspect type data from the meta data and creating the asset data model automatically from the aspect type and the asset type by mapping corresponding data, wherein the aspect type data are determined at least from the meta data, structure definition data, structure field data and field meta data, and the asset type data are deter mined at least from the aspect type data,

c. generating 130 the control message, comprising the asset data model, d. publishing 140 the control message using the publish-subscribe communication principle to the IoT-device.

The asset data model is automatically created by data mapping.

The meta data comprise structure definition data, structure field data and field meta data.

The aspect type data are determined at least from the meta data.

The asset type data are determined at least from the aspect type data.

In software architecture, publish-subscribe is a messaging pattern where senders of mes sages, called publishers, do not program the messages to be sent directly to specific re ceivers, called subscribers, but instead categorize published messages into classes without knowledge of which subscribers, if any, there may be. Similarly, subscriber express inter est in one or more classes and only receive messages that are of interest, without knowledge of which publishers, if any, there are.

Publish-subscribe is a sibling of the message queue paradigm and is typically one part of a larger message-oriented middleware system. Most messaging systems support both the pub/ sub and message queue models in their API, e.g. Java Message Service (JMS). This pattern provides greater network scalability and a more dynamic network topology, with a resulting decreased flexibility to modify the publisher and the structure of the pub lished data.

In other words, publishing the control message using the publish-subscribe communica tion principle to the IoT-device means using the pub/ sub principle for the communication of the control message for the distribution of the scope of the initial message to IoT- device.

Fig. 2 depicts an example of data structures used for the invention. For instance, metadata can contain:

• DataTypeSchemaHeader: array of definition of structuredDataType

• FieldMetadata: array of structuredDataType instances

An asset is a digital representation of a machine or of an automation system with one or more automation devices connected to the Siemens MindSphere® platform (e.g. PLC). Data collection and data delivery with MindSphere relies on so-called (virtual) assets. An asset can be any element, e.g. as a pump, a motor, a PLC, a complete machine tool, a pro duction line, a robot, a crane, a vehicle, a wind turbine, etc., just to name a few examples. The data describing an asset is collected and transmitted to MindSphere. These data are then made available for further processing and evaluation.

Aspects are data modeling mechanisms for assets. Aspects group related data points based on their logical association. Example: The pump set has an aspect, for example, "Energy Consumption", which contains the data points "Power", "Voltage" and so on. The aspect should have a connection to data points and a physical asset. An aspect can com prise several variables

A Type is a preconfigured template for an asset. Assets assume the properties of the type on which they are based. Within the Type you can define which Aspects are integrated into the template.

AspectType 311 and AssetType 301 are templates for Aspects and Assets, i.e. templates for a predefined structure.

Thus, AspectType 200, 201 are mapped to a DataPointAssetType 210.

Based on a specific custom object, AssetTypes and Assetlnstances are created by using raw values from the definition of structuredDataTypes, for instance according to Fig. 2. The aspect type 200, 201, 311 and an asset type 301 are used in step b. of the aforesaid method.

Fig. 3 shows an example of meta data used at the invention.

Fig. 4 shows an example of a domain model of AssetType 301 - AspectType 311 and re spective instances. An Assetlnstance 300 defines an AssetType 301. One or more elements of Aspectlnstance 310 define a respective AspectType 311.

Fig. 5 shows an embodiment of the AspectType 311 for SPSFields for the given example of MetaData according to Fig. 3.

Fig. 6 illustrates a flowchart 400 with an example of automatic creation of Asset- AspectTypes corresponding to a system arrangement of Fig. 7, which shows as an embod iment according to the invention embedded in an IoT system based on the platform 501 MindSphere® by Siemens. Not all details of the data flow are shown for the sake of better understanding.

The MindSphere® platform 501 for automatic creating an asset data model using a pub- lish-subscribe software architecture, and for automatic generating and publishing mes sages based on the asset data model, comprises means configured to carry out the afore said method.

A system according to the invention comprises the aforesaid platform 501 and one or even a plurality of connected IoT-devices. A messenger is configured to control at least one of the plurality of IoT-devices by at least one published message.

A Publisher 410 (acts as a message input) as an MindSphere-input provides messages to a MQTT Broker 420.

The initial message comprises for instance MetaData 411, keys, delta information, status information regarding events or "Keep Alive".

MQTT (Message Queuing Telemetry Transport) is an ISO standard (ISO/IEC PRF 20922) publish-subscribe-based messaging protocol. It works on top of the TCP/IP protocol. It is designed for connections with remote locations where a "small code footprint" is required or the network bandwidth is limited. The publish-subscribe messaging pattern requires a message broker. The MQTT Broker 420 approves the validity by an Authentication Server 510 via a JWT Validation 422 (JSON Web Token Validation) and returns an "Allow"-Signal 511, if au thenticated positively.

In computing, JavaScript Object Notation (JSON) is an open-standard file format that uses human-readable text to transmit data objects consisting of attribute-value pairs and array data types (or any other serializable value). It is a very common data format used for asynchronous browser-server communication.

A JSON Web Token (JWT) is a JSON-based open standard (RFC 7519) for creating access tokens that assert some number of claims. For example, a server could generate a token that has the claim "logged in as admin" and provide that to a client. The client could then use that token to prove that it is logged in as admin. The tokens are signed by one party's private key (usually the server's), so that both parties (the other already being, by some suitable and trustworthy means, in possession of the corresponding public key) are being able to verify that the token is legitimate. The tokens are designed to be compact, safe and usable especially in a web-browser single-sign-on (SSO) context. JWT claims can be typi cally used to pass identity of authenticated users between an identity provider and a ser vice provider, or any other type of claims as required by business processes.

JWT relies on other JSON-based standards: JWS (JSON Web Signature) RFC 7515 and JWE (JSON Web Encryption) RFC 7516.

The Authentication Server 510 can be implemented as a Redis (Remote Dictionary Server), which is an open source in-memory data structure project implementing a distributed, in memory key- value database with optional durability. Redis supports different kinds of abstract data structures, such as strings, lists, maps, sets, sorted sets, hyperloglogs, bit maps, streams and spatial indexes.

A Subscriber 430 fetches 431 the validated messaged 421 in form of encoded MetaData and forwards them via SQS 502 to an UADP Parser 440 (Unified Access Data Plane Par ser), and/ or the further parsers, e.g. a JSON Parser 520 (JavaScript Object Notation Par ser), which is connected via SQS 502 to the Subscriber 430 in order to enhance the instruc tion set of the Mindsphere platform 501. A binary data format can be used for a simple parser as well.

Amazon Simple Queue Service 502 (SQS) is a fully managed message queuing service that enables you to decouple and scale microservices, distributed systems, and serverless ap- plications. SQS 502 eliminates the complexity and overhead associated with managing and operating message-oriented-middleware and empowers developers to focus on dif ferentiating work. Using SQS 502, you can send, store, and receive messages between software components at any volume, without losing messages or requiring other services to be available.

SQS 502 offers two types of message queues. Standard queues offer maximum through put, best-effort ordering, and at-least-once delivery. SQS 502 FIFO (first-in first-out) queues are designed to guarantee that messages are processed exactly once, in the exact order that they are sent.

Unified Access Data Plane (UADP) is a standardized technology for fixed and wireless networking.

The UADP Parser 440 is responsible of decoding 441 OPC UA Pub/ Sub messages, which can be for instance Discovery messages (contains Metadata). Moreover, the UADP Par ser 440 is also responsible of extracting raw values from decoded metadata to build a cus tom object 442, containing a respective Assetlnstance 300 and AssetType 301.

The UADP Parser 440 comprises in this example a Redis 505 as well as a RDS 506 (Rela tional Database Service).

The UADP Parser 440 generates in this embodiment two data paths, by which data are sent.

In a first path the UADP Parser 440 is connected to an IoT Writer 530 via SQS 502 and comprises timeseries data 540, which are dynamic data, e.g. a data set with a value at a certain point of time and the time stamp from that point of time. In this example the IoT Writer 530 has an implemented RDS 506. The IoT Writer 530 checks the created asset in stances and stores the timeseries data 540 under the specific asset instance.

In a second path the UADP Parser 440 is connected to an AssetWriter 450 via SQS 502, which transfers a custom object 442 containing Assetlnstance 300 and AssetType 301. In this example the AssetWriter 450 has an implemented Redis 505.

The Redis 505 and the RDS 506 are embodiments for storing data, but other implementa tions as applicable as well, although the dedicated implementations of Redis 505 and RDS 506 offer performance benefits. The OPC UA PubSub API (Application Program Interface) comprises several micro ser vices. One of those micro services is the AssetWriter 450, which is responsible for creating the MindSphere ® IoT model and map the OPC UA PubSub data to that model.

The data mapping is performed by step b. of the aforesaid method:

b. creating 120 an aspect type 200, 201, 311 with aspect type data and an asset type 210, 301 with aspect type data from the meta data and creating the asset data model automatically from the aspect type 200, 201, 311 and the asset type 210, 301, wherein the aspect type data are determined at least from the meta data, structure definition data, structure field data and field meta data, and the asset type data are determined at least from the aspect type data,

The AssetWriter 450 creates the AspectType for a "fields" field in the StructureDefinition, which contains all defined fields in StructureFields. Subsequently the AssetWriter 450 creates 451 the AssetType 210, 301 containing the AspectType 200, 201, 311. Acordingly, the AssetWriter 450 creates 452 the Assetlnstance 300. Both creation steps 451, 452 can be performed by REST (REpresentational State Transfer) and constitute data for a subsequent AssetManagement unit 460, e.g. in form of a Fleet Manager 560, which manages connect ed IoT devices.

The AssetWriter 450 creates Assetlnstance 300 and AssetType 301 using raw values from the definition of structuredDataTypes according to Fig. 2.

The two paths are used for modeling data and to store the timeseries data 540.

The IoT Writer 530 forwards data to the AssetWriter 450 for modeling the timeseries data 540.

The timeseries data 540 constitute data for subsequent control of connected IoT devices, e.g. in form of an Energy Management (EM) application program 550.

After type and instance creation and subsequent mapping, i.e. MetaData is processed, the timeseries data can be uploaded as DataSet messages into the Mindsphere platform at the RDS 506.

The mapping according to Fig. 1 and 2 is performed at the AssetWriter 450.

The IoT Writer 530 is responsible to the handling of the DataSet messages and stores the timeseries data 540 under relevant asset instances at the RDS 505. The AssetWriter 450 is responsible to keep and to store that information at the Redis 505. Not only Asset and Aspect types are created but also Asset Instances are created. When ever timeseries data 540 arrive to the MindSphere platform 501, the Assetlnstance 300 owning the relevant data is detected and all timeseries data 540 under that specific As setlnstance 300 is stored.

List of reference numerals:

100 Process

110 Receiving data

120 Creating aspect/ aspect type

130 Providing data

140 Generating and publishing data, optionally present aspect/ aspect type

200, 201, 311 Aspect Type

210 Data Point Asset Type

300 Asset Instance

301 Asset Type

310 Aspect Instance

400 Flowchart

410 Publisher

411 MetaData (encoded)

420 MQTT Broker (Message Queuing Telemetry Transport Broker)

421 MetaData (encoded)

422 JWT Validation (JSON Web Token Validation)

430 Subscriber

431 fetch MetaData (encoded), by SQS

440 UADP Parser (Unified Access Data Plane Parser)

441 Decode MetaData

442 Custom Object containing Asset Types and Instances, by SQS

450 AssetWriter

451 Create AssetType

452 Create Assetlnstance

460 AssetManagement

500 OPC UA System (Open Platform Communications Unified Architecture

System)

501 MindSphere IoT Environment

502 SQS (Amazon Simple Queue Service)

505 Redis (Cache, in-memory data base))

506 RDS (Relational Database Service)

510 Authentication Server

511 Allow

520 JSON Parser (JavaScript Object Notation Parser)

530 IoT Writer 540 IoT TimeSeries

550 Energy Management Program "EM App"

560 Fleet Manager