AC IoT & Industry 4.0
The application center “IoT and Industry 4.0” at DAI-Labor focuses on the digitalization and scalable interconnection of devices, services, and human actors. It brings together the research conducted in several CCs at DAI-Labor under the concept of the Internet of Things (IoT) with a focus on future industrial systems.
Following the global interconnection of people and services through the Internet, the IoT concept has been on the rise with the ambition to digitalize and interconnect everyday objects in many domains of life and work. The great potential of this ambition lies in the holistic networking of the involved entities, i.e., the ability of any device to interact with any other and enrich their collective functionality and benefit.
On the other hand, the increasing number and variety of devices or sensors in the growing IoT world makes it ever more complex to realize and manage such holistic interconnection systems.
Through our research projects with industry partnerships, we are trying to tackle such problems and realize the dynamic scaling and adaptability features of distributed IoT systems, which are particularly required in Industry 4.0 scenarios.
As the digitalization trend goes on with an increasing pace and with the involvement of a diverse set of actors, proper management of this digital layer as well as the services deployed over it becomes ever more crucial. Research and development at DAI-Labor on the IoT middleware technologies addresses this need for efficiently and scalably bridging the heterogeneous set of devices and their data with the dynamic scene of novel applications and scenarios.
Machine learning is a part of artificial intelligence to recognize patterns and laws on the basis of large amounts of data and intelligent algorithms. Our team mainly deals with Time Series Mining, Deep Learning and Automated Machine Learning. In Automated Machine Learning, we are examining and developing methods for automating algorithm selection and hyper parameter optimization. The goal is to make the complex nature of applying machine learning methods more accessible also for non-experts. We develop Deep Learning architectures of artificial neural networks capable of learning representations, concepts and abstractions from complex data applied on industrial application problems in the context of big data. For this, we consider multi-layer perceptrons, convolutional networks, autoencoders, Boltzman machines, and recurrent networks as basic models. In Time Series Mining, we aim at developing a sound theoretical foundation for a predominalty application-dominated field in order to derive improved classification and clustering methods.
Our research focuses on developing human-aware cognitive models for robots in order to provide adaptive personalized assistance to humans in work environments or at homes. At the low-level of such models, we focus on robot cognition abilities, such as human presence detection, hand and head gesture recognition, attention detection and action recognition in the context. Whereas at the high-level, these low-level signals are used to anticipate human mental states and intentions to better understand their needs for more reliable, natural and adaptive collaboration. This is done through our novel stochastic decision-making and policy selection algorithms, adapting to a human’s changing contextual behaviors, preferences, and characteristics like their collaborated task skills.