CASUS - Center for Advanced Systems Understanding
CASUS combines methods from mathematics, systems theory, data science and scientific computing at a single location, to rethink data-intensive systems research across individual disciplines. CASUS is under development as a new research institute in Görlitz during 2019-2022 and will reach full strength in 2023.
In the future, understanding complex, networked systems will make a decisive contribution to mastering society's important challenges. For the first time in human history, novel digital methods will enable us to understand this complexity and master it by networking various scientific disciplines.
CASUS is planned as the centre for digital interdisciplinary systems research in Germany and is to occupy a leading international position in this up-and-coming field of research.
Digital interdisciplinary systems-research researches and develops the latest and most innovative methods from mathematics, modelling, simulation, data and computer science to solve questions from such diverse areas of systems research as earth system research, systems biology or materials research.
CASUS aims to bring together the best scientists from these fields at a joint institute in order to develop visionary ideas in interdisciplinary teams on how to master the complex challenges of the future with digital methods.
CASUS is to become the centre for digital interdisciplinary systems research in Germany. CASUS aims to create digital, dynamic "worldviews" of complex systems that combine large amounts of data about these systems with novel methods of modelling such systems in order to create a digital image of complex reality based on systems and their interactions and thus be able to make predictions. The understanding and predictability of the development of complex systems will become increasingly important in the coming years, e.g. for a better understanding of the development of complex organisms, the long-term development of the Earth system and the development of novel materials, and thus become increasingly important for research as well as for business and decision makers.
The underlying assumption of CASUS is that in the future this knowledge and understanding of the complexity and diversity of the world will experience disruptive change through the use of novel digital methods from big data and large-scale simulations. Systems research will play a central role in this, in order to conduct joint methodological research across scientific disciplines. CASUS will rethink the use of state-of-the-art technologies and methods. These should make the best possible use of existing technologies, decisively advance systems research with novel algorithms and design these methods in such a way that they are available to the broadest possible circle of scientists without specialist knowledge.
An institute with this orientation does not yet exist, which is why CASUS should become an attractive location for internationally leading experts who come from the various disciplines of systems research and its applications, method research for modelling and data analysis as well as mathematics and work together in interdisciplinary teams. CASUS aims to bring together the best minds for cutting-edge research in digital systems science in one place. In doing so, great importance is attached to innovative and unorthodox research approaches in order to overcome historically grown structures of individual disciplines and to promote interdisciplinary solutions. An essential part of the CASUS concept is an attractive international fellowship and workshop program for top international scientists.
Research Focus: Matter Under Extreme Conditions
Modern materials research investigates the relationships between the structural and functional properties of materials and the structure and dynamics of materials at the atomic level, which are dominated by quantum mechanical processes. Its aim is to tailor novel materials and to understand the dynamic behaviour of these materials based on the understanding and control of atomic processes. Research on matter under extreme conditions such as high fields, densities and temperatures is an extension of classical materials research and allows links, for example, to energy research, plasma physics and astrophysics in the laboratory. Here, dynamic properties of matter and non-equilibrium states become significantly more important and are needed to understand new states of matter.
Scientists at CASUS investigate the non-equilibrium behaviour of matter under the influence of extreme electromagnetic fields, temperatures and pressures. Understanding these exotic states of matter will contribute to:
- fundamental understanding of strongly correlated quantum systems
- knowledge of planetary and stellar interiors
- understanding of astrophysical processes
- development of plasma-based compact particle accelerators for applications such as tumor radiation therapy or compact X-ray light sources
Research Focus: Earth System Research
Research into the entire Earth system must be performed in order to understand the connections between geology, climate, ecology and human influence in the interplay of these complex systems. In this context, an ever-increasing amount of sensor data has to be brought together in real time and compared with models. Improving the predictive power of these models will influence political decisions, economic developments and people's daily lives.
Data- and computation-intensive computer models are developed at CASUS that allow the ecological, hydrological and economic effects of global change to be studied in high spatial and temporal resolution and in their complex interactions.
CASUS will provide predictions for entire ecosystems and their ecosystem functions for the next 50 to 100 years, including their:
- biogeochemical cycles
- water quality and quantity
- biomass production and agricultural yields
allowing the quantitative analysis of the overall diversity of event chains and of feedback mechanisms between them, including the overall economic costs and ecological consequences.
Research Focus: Systems Biology
In systems biology the quantum mechanical and quantum chemical nature of the molecules that make up the individual cells, proteins and other functional components of organisms is to be linked to the development of the entire organism. In addition, data must be obtained in vivo, i.e. on the living organism, which is constantly changing and constantly in disequilibrium. In addition, the subsystems of an organism are networked in such a way that they can only be understood in interaction. Huge amounts of data from imaging processes, genetics, biochemistry and modelling must be combined and integrated into a common picture of the origin and function of organisms.
CASUS performs basic research to enable next-generation biology, personalized medicine, and mechanistic understanding and control of living systems. This includes:
- virtual and augmented reality for the laboratory of the future
- computational prediction and control of biological processes
- learning and inference of computable models from microscopy data
- biomedical data science and AI
CASUS’ methods and algorithms will allow us to deeply understand living matter and the processes of life in their inner function across multiple scales, from molecules to tissues, based on terabytes per day of data obtained from high-resolution 3D microscopes.
Research Focus: Autonomous Vehicles
In addition to a comprehensive digital representation of the environment, automated decisions require networking with the surrounding infrastructure and betweenindividual road users. This results in a highly dynamic system whose real-time analysis will make a decisive contribution to the safety of autonomous vehicles. The major challenge is to interpret and correctly predict the behaviour of non-networked actors such as humans and animals in order to include them in automated decision-making. The approach pursued by CASUS to develop digital images with a high degree of reality and high predictive power is thus becoming both important for our future everyday life and a direct reference to technological innovation in the economy.
CASUS’ use of machine learning methods in combination with behavior models and sensor data will create a digital image of the current surroundings for:
- navigating the vehicle
- making decisions about the behavior of other road users
- communicating with other road users and existing infrastructure
- the vehicle’s power consumption