emergenCITY

The LOEWE Focus Project NICER, which expired at the end of 2018, resulted in a project proposal for the LOEWE Center emergenCITY, aiming at achieving resilience for the digital city. The application submitted in December 2018 was approved in June 2019. The official project start was on January 1, 2020. The same universities that partook in NICER (TU Darmstadt, University of Kassel and University of Marburg) are represented in the new Center, but the number of research groups involved has roughly doubled. The universities in Kassel and Marburg now contribute with two research groups each.

The LOEWE center emergenCITY is divided into four project areas: Cyber-Physical Systems, Communications, Information Systems as well as City and Society. The emergenCITY Laboratory, supported by all research groups, complements the four project areas.

The work package of the Distributed Systems Group belongs to the area Cyber-Physical Systems. The title is "Situational Cooperation of Cyber-Physical Agents for Resilient Urban Mobility". The elimination of central information and traffic management systems in the event of a crisis raises the urgent research question of how a large number of (partially) autonomous, heterogeneous, mobile agents (e.g. transport vehicles, rescue robots, drones, software agents) can maintain emergency logistics services through decentralized, self-organized cooperation among themselves.

This research is carried out in close cooperation with our project partners of the Institute of Flight Systems and Automatic Control (FSR) at TU Darmstadt as well as with the other research projects   in the emergenCITY framework. Our work program is structured into four activities:

(1)    What information is required to set up logistics services for an emergency operation? This information is made available in a common world model, whereby various forms of storing the world model in the agents must also be examined.

(2)    Then, available, spatially distributed information sources (e.g. onboard sensors, networked agents, merged data sources, user input) need to be identified and configured in order to compensate for any failed information systems.

(3)    Building on this, the required software architecture and implementation of a cooperation platform for multi-agent scenarios are explored. This includes semantic modeling, i.e. a common understanding of the elementary basic skills of the agents that they may contribute to the multi-agent system in order to achieve a common goal. In addition, operational plans for the cooperative activities must be designed.

(4)    Finally, the results of the previous three activities will be used to develop decentralized, probabilistic, cooperative processes that enable agents to design their own trajectories without conflicts.

Further information on emergenCITY can be found at: www.emergenCITY.de.

Contacts: Yasin Alhamwy and Prof. Kurt Geihs.