Internet of Things
Service-based revenue models build upon data generated from embedded IoT devices to enable predictive maintenance of machinery, for example, or to suggest process refinements based on usage patterns. Manufacturers in the industrial sector are now pushing the envelope with AI and ML, leveraging edge capabilities to drive competitive advantage through streamlined operations and processes. Developers of service-based revenue and
T
Product-as-a-Service, (PaaS) models, facing the twin challenges of speed to market and reduced development costs, need rapid access to structured data, and middleware is emerging as a valuable tool. By simplifying the interconnection of multiple, disparate IoT platforms, and providing a layer of abstraction, middleware, such as Clea from SECO, allows the developer to concentrate on creating value- adding applications.
THE EVOLUTION OF THE IIOT Thanks to the extensive array of sensors and embedded devices integrated into all types of modern equipment, rich data sets are available which can give detailed insights into the operation of the equipment in question. At the same time, advances in the processing power of embedded IoT technology enable sophisticated algorithms, including AI applications, to be hosted at the edge, devolving decision-making from the cloud, and reducing latencies where real-time responses are required. IoT applications are therefore becoming progressively data-driven, as developers focus on creating applications which derive value from the data. OEMs are increasingly leveraging AI-based models to develop services linked to capital items, thereby monetising the data streams. Unfortunately, due to the rapid pace of innovation, and despite standards such as SMARC and COM Express in the hardware world, IoT architectures are extremely heterogeneous, complicating the developer's challenge. The sheer amount of knowledge required to integrate a wide range of disparate platforms, figure 1, while also deploying AI-based solutions requires a team, rather than an individual, increasing both development costs and time.
Developers are therefore looking towards tools such as middleware, which introduce a layer of abstraction over the “plumbing” and enable them to concentrate on building AI- based applications.
Middleware software simplifies the connection of applications, services, and data sources that were not originally designed to interact with one another, shielding developers from the intricacies of underlying hardware platforms, operating systems, and network protocols. By minimising duplication of efforts and promoting collaboration between applications, middleware
56
he ever-evolving IoT is entering a new phase, where OEMs and manufacturers are focusing on creating value through the servitisation of their products.
CREATING AI- BASED SERVICE REVENUE MODELS
By Francesco Vaiani, software product manager at SECO
Fig. 1: IoT architectures are extremely heterogeneous
https://ietresearch.onlinelibrary.wiley.com/doi/full/1 0.1049/cim2.12077
optimises the overall efficiency of the IoT ecosystem, streamlining the development process and freeing up valuable development time and resources.
Clea is an example of a powerful software platform which integrates data orchestration with its core middleware components and also supports rapid deployment of AI models at scale.
SECO’S CLEA SOLUTION
The open source Clea software suite gives the developer full control over the data journey, from how it is collected to how it is processed to where it is stored. This modular software stack scales from POC to full production deployments involving huge numbers of devices and exchanged messages. Built on Kubernetes, Clea is cloud-agnostic and provides developers with all the tools required to orchestrate data in the Cloud and clean it before adding intelligence. Clea abstracts the developer from the complexities of the underlying IoT platforms and offers device management features, including secure OTA updates of firmware, operating systems, and containerised applications running on IoT devices.
The solution, which can run Google Cloud Services out-of-the-box also facilitates the rapid development of value-added services, including advanced AI applications. As well as providing developers with all the tools required to build customised AI applications, Clea also offers a number of pre-built structures through the Clea App Framework.
The Clea suite consists of three core components, figure 2, which communicate to offer all the advantages of a complete, modular, and scalable solution. Portal is the user-facing part of Clea, providing
an extensible IoT front-end facilitating service monetisation by integrating a billing framework that allows reselling edge or cloud applications within a subscription model. As a web-based front-end, Portal enables managed user access, device and data visualisation, reporting, and fine-grained permissions. Portal’s user interface is highly customisable, and its application framework allows developers to extend the front end with custom web applications that can interact with edge devices through Astarte (Clea’s device-cloud data hub) and Edgehog (Clea’s device and fleet manager) APIs. Edgehog is an open-source device agnostic IoT platform which simplifies IoT device and fleet management throughout the lifecycle. Built upon Astarte, Edgehog is highly scalable, from individual devices to network-wide fleets and supports fundamental operations, including deployment of update campaigns, device status and connection information monitoring, geolocation, and file transfer.
Astarte is a comprehensive, open source, IoT middleware, containing the tools and functionality for orchestrating and modeling data, and handling communication between devices and the cloud. Astarte manages any aspect of field data and enables advanced use cases such as data reduction. Its off-the-shelf SDKs allow integration within any edge device, including microcontroller-based systems, with no development effort needed. Astarte features an extensive set of APIs and integrations, making it straightforward to build applications and integrate with third-party frameworks. SECO is investing heavily in making Google Marketplace a preferred distribution channel for Astarte. Google Cloud Marketplace lets users quickly deploy functional software packages that run on Google Cloud, allowing customers to easily start up a familiar software package with services like Compute Engine or Cloud Storage, with no manual configuration required. This level of integration with a trusted, global cloud provider gives the customer one-click access to fully-fledged, managed software installations, such as Astarte, simplifying delivery, improving the consistency and efficiency of deployment, and reducing cost. The expansion of cloud provider marketplace solutions enables the developer of complex applications to source software from a variety of vendors while encouraging an open technology environment. Following Astarte, the other two modules of the suite – the Edgehog device manager and
May 2024 Instrumentation Monthly
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84
