How Digital Twins foster innovation in IoT-enabled environments
By Matthew Mikell | 4 minute read | February 6, 2018
Dynamic digital representations, or Digital Twins, are rapidly changing the way industries design, build and operate their products and processes. Gartner predicts, “by 2021, half of large industrial companies will use Digital Twins, resulting in those organizations gaining a 10 percent improvement in effectiveness.”
Powered by the Cloud, IoT, and AI, Digital Twins enrich complex systems like cars, wind turbines and buildings across their entire life cycles. A Digital Twin combines design, production and operational data. It allows assets to be tested before, during and after production, and across a wide range of environments.
IBM Research – Ireland is developing different Digital Twin technologies. These include:
- A virtual platform for testing of complex IoT systems with live and simulated data.
- Forming a knowledge graph for IoT that combines reasoning with machine learning to allow the system to autonomously analyze and understand life cycle data.
Utilizing a virtual testing platform for IoT systems
In order to test complex IoT Systems, our researchers are using a virtual platform. This allows designers and developers of transportation services to investigate large-scale connected car services. They achieve this by merging simulations of large-scale automotive IoT deployments with proof-of-concept capabilities provided from real world vehicles. This platform is helping automotive partners design their services at scale while accelerating time to market.
The platform also allows drivers of actual vehicles to experience a large-scale connected scenario first hand. This combination of simulated and real-world data generates valuable insights. These insights are critical to user-centric development, resulting in reliable systems that are ready for the market. By embedding the data from actual vehicles into the digital environment, we can test the effects of assisted and autonomous driving in large-scale traffic simulations, in real time.
For example …
In collaboration with University College Dublin (UCD), we are using our virtual testing platform to evaluate a number of new mobility concepts. For example, we are testing a new car sharing mobility service that dynamically adapts to user preferences. This then allows a group of users to meet based on changeable traffic conditions and their variable pick-up time arrangements.
We are also investigating using IoT services to maximize air quality intake for pedestrians and cyclists by reducing their exposure to pollution. Imagine an electric bike using IoT devices, such as mobile phones and sensors. These IoT devices detect and automatically assist the cyclists when traveling through areas of high pollution. In those areas, the engine of the e-bike would be automatically triggered into operation. When that happens, it reduces the cyclist’s pedaling effort, resulting in a lower breathing rate and lower pollution intake. The virtual testing platform can also be used to connect to the e-bike and monitor how the cyclist would actually react to this new service, investigating the interactions between the cyclist and the bike.
Another service solution we are evaluating would reduce a pedestrian’s exposure to car exhaust pollution. How? The AI controls of a hybrid car to automatically switch between combustion and electric mode when the vehicle is in close proximity to pedestrians and cyclists.
These examples illustrate how a virtual testing platform can help accelerate the development of new services. At the same time, it also helps the transportation industry respond to the ever-increasing demands for environmental accountability.
Automating Insights with a knowledge graph for IoT
At IBM Research – Ireland, we are developing AI technologies to connect and understand IoT data in new ways. We’re combining machine learning with knowledge graph reasoning to enhance data being extracted from an IoT network. And we’re also adding layers of semantic meaning to create new insights within the network. This technology is the Digital Thread at the core of each Digital Twin. It connects information along the lifecycle stages into a knowledge graph. This graph then enables new informed decisions and automation of processes.
By using a knowledge graph, we are able to organize data and its variables being extracted into groups and establish the relationships between the data sets and their variables. The knowledge graph provides a shared vocabulary of information that can be used to create a model of a domain, the types of data within it, their properties and the relationships between the data–and we are using natural language to do all of this.
As a result, our AI solution understands the meaning and the relationships between the different types of data within a network or system. This gives our research teams new ways to derive innovative insights from an IoT system and present them as new knowledge and information to end users.
For example, take an IoT temperature sensor in a building. The temperature sensor has data readings, the type of data that it is recording and its location. Our AI system understands general concepts of physics and how temperature is influenced by heating or cooling, such as environmental factors, heat system controls and so on. This allows our system to form a knowledge graph to understand the temperature settings within the building and the multiple factors that impact the temperature within its operating environment. This allows for the self-diagnosis of problems within the system while enabling it to learn and understand this relationship over time. It is also scalable and works across industries such as retail and automotive.
Our virtual testing platform and knowledge graph for IoT demonstrate the value of Digital Twin. We’re enabling industries to create better informed designs, optimize production, and manage efficient operation. The virtual testing platform can simulate these large-scale environments and networks while providing a way to perform controlled user-acceptance tests.
This combination of simulated and real-world data generates valuable insights that are critical to systems development. Our knowledge graph for IoT is a scalable solution that enables IoT to learn system behaviors, to understand management operations and to self-diagnose problems. And all while making human-machine interaction more natural and intuitive.
We will demonstrate the knowledge graph for IoT at the IEEE flagship IoT conference World Forum IoT, February 5-8th in Singapore. A prototype of the virtual testing platform will be shown at the ENABLE-S3 consortium General Assembly, Review and Marketplace event. This is scheduled at our Research lab in Dublin on July 4, 2018.
For deeper research on Digital Twins and related topics, see:
Joern Ploennigs, Amadou Ba, Michael Barry, Materializing the Promises of Cognitive IoT: How Cognitive Buildings are Shaping the Way, IEEE Internet of Things Journal, 2017 Wynita Griggs, Giovanni Russo, Robert Shorten, “Leader and Leaderless Multi-Layer Consensus With State Obfuscation: An Application to Distributed Speed Advisory Systems”, IEEE Transactions on Intelligent Transportation Systems, 2017
The ENABLE-S3 project has received funding from the ECSEL Joint Undertaking under grant agreement No 692455. This Joint Undertaking receives support from the European Union’s Horizon 2020 research and innovation programme and Austria, Denmark, Germany, Finland, Czech Republic, Italy, Spain, Portugal, Poland, Ireland, Belgium, France, Netherlands, United Kingdom, Slovakia, Norway.