This is what the future could be like: A company directly incorporates a sensor unit that analyzes data and detects anomalies into each installation it produces. The installations are sold throughout the world, and as soon as they are in operation they transmit their data to a common cloud. In this way all installations throughout the world can learn from each other how normal operation should proceed. If there is a deviation this is recognized, even if this unknown pattern has never occurred in the individual installation. "What's special about this is that we can react to operating conditions that have never yet arisen before and have a system that is constantly evolving. Through learning by itself it recognizes normal operating conditions and deviations" explains Dr. Mario Aehnelt, Head of the Department "Visual Assistance Technologies" at the Fraunhofer IGD in Rostock.
Optimum algorithm incorporation for every customer
Data@Hand is an information and data tool for humans in working processes which is aimed at process optimization and is based on the principles of machine learning and artificial intelligence. It supports the analysis of complex data volumes but leaves the specific decisions on how to react to anomalies to the professional expert. Through individually addressed questions Data@Hand ensures optimum algorithm incorporation for every customer. In the same way as, for example, vital data of a patient, machine data from production can be evaluated more quickly. Analysis can not only take place via a powerful server-based platform, but also on ultra-small systems directly at the machine or patient.
Data@Hand can also be connected to existing AI tools and data processing platforms (MES/ERP) or be used for visual data formatting, by way of Plant@Hand3D or Health@Hand, for example. Customers can therefore work using systems they are already familiar with.
Live data analysis at the Hanover Trade Fair
At the Hanover Trade Fair scientists from the Fraunhofer IGD will show how a real additional value can be generated from a pure data collection through intelligent analysis with Data@Hand and the visualization of critical conditions. In an example demonstration the operating conditions in a compressor unit will be modified to different degrees and the machine parameters of temperature, vibration and power uptake will be analyzed. The recognition analysis runs on the directly connected sensor unit. With these data, anomalies and new operating influences are identified in real time. As soon as the operating behavior deviates from normal a warning is given. With the obtained data, not only the causes of problems can be analyzed, but it is also possible to predict what contributes to reducing maintenance costs.
At the 2019 Hanover Trade Fair the technology can be tested live on the trade fair exhibit from April 1 to 5, 2019 at the joint Fraunhofer booth A30 in Hall 6.
Many people use Alexa, Siri and other similar voice assistants on a daily basis, dipping in to access the latest news, make use of voice navigation or simply stream their favor-ite songs. Voice assistants are an intuitive way to interact with technology, an effective way of delivering services and imparting information. They are not just handy everyday helpers, however; they present companies and business with a huge opportunity to simplify human-machine interaction and offer entirely new services to their industry customers.
Focus on companies
Researchers at Fraunhofer IAIS in Sankt Augustin develop just these sorts of voice inter-action systems for use in a wide variety of applications, including manufacturing and the automotive and medical sectors. While Alexa, Siri and the like are aimed at indi- vidual consumers, the research team at Fraunhofer IAIS uses the latest techniques in machine learning, question answering and smart graphical knowledge networking to address the specific needs and challenges of business. “In manufacturing, for instance, we are seeing more and more robots equipped with voice assistants, which the worker can then operate and train using voice and gestures,” says Prof. Dr. Jens Lehmann, Lead Scientist at Fraunhofer IAIS.
Prof. Lehmann and his team at Fraunhofer IAIS specialize in dialog systems catering to domain-specific knowledge and trained for specific applications. At the Hannover Messe, they will be showcasing a voice assistant integrated into a VW Tiguan. Wearing a headset and virtual reality glasses, drivers will be taken on a virtual tour of Berlin while the interactive system answers questions about the surroundings such as: What’s
RESEARCH NEWS
RESEARCH NEWS
March 1, 2019 || Page 2 | 3
that building on the left-hand side? What’s it known for? When was it built? Who built it? The system also supports supplementary questions such as “Where does the archi-tect come from?” or “Tell me more about him!”
Domain-specific knowledge enables complex questioning
The Hannover Messe showcase is a collaboration between the Fraunhofer Cluster of Excellence Cognitive Internet Technologies (www.cit.fraunhofer.de), Volkswagen and the Fraunhofer Institute for Integrated Circuits IIS. “All the knowledge relating to Berlin has been collated into a knowledge graph, where each building represents a point on the graph and forms connections with other points. As a result, we can gather progres-sively more information and constantly expand the knowledge base. This is what allows for complex questioning instead of restricting inquiries to a limited number of pre-scribed questions,” explains Lehmann. In a manufacturing context, this sort of knowledge graph could report on the status of machines, for example, or answer ques-tions about components produced in the last hour. The knowledge graphs used for the trade show exhibition draw on a variety of data sources including DBpedia and Open-StreetMap. A special feature of the voice assistant is that it is also able to harness un-structured knowledge, such as text documents on museums, for instance.
Cyber-physical systems are a key Industrie 4.0 technology and represent another poten-tial area of application. With these systems, you have not only the physical machine in the production hall, but also a virtual counterpart that is fed with real data. This data can be interrogated using dialog or question answering systems. “While question an-swering systems are simply there to answer individual questions, dialog systems sup-port multi-layer dialog combining sequences of questions and answers. A dialog system will also respond to sequences of inquiries and small talk, just like the exhibit we will have on display,” says Lehmann.
The more training data, the smarter the voice assistant
“It is the domain-specific knowledge that makes a voice assistant smart. The technical challenge from our side lies in developing a system that can understand users’ queries and respond appropriately using the knowledge contained in the knowledge graph,” the researcher concludes. Developing such a system calls for the application of the lat-est techniques in machine learning, techniques that the researchers at Fraunhofer IAIS are constantly developing and refining. The expertise they have assembled in machine learning and domain-specific knowledge puts them at the top of their field internation-ally. Tailored to the respective domains, the experts select the appropriate machine learning algorithms and train them using sample dialogs and question-answer pairs. The intelligence of the voice assistant grows with the amount of training data it amasses. The voice assistants developed by Fraunhofer IAIS offer their users the ulti-mate digital experience and are all GDPR-compliant. Visitors can try out the exhibition demonstrator live at the Hannover Messe from April 1 to 5 at the joint Fraunhofer Booth C22 in Hall 2.
Fraunhofer-Gesellschaft