Projects

The modern computing landscape ranges across a large variety of heterogeneous device types with different amounts of available resources; starting with microcontrollers and mobile devices, multicore desktop computers with accelerator hardware such as graphics cards to dedicated server clusters. In order to alleviate specific resource restrictions, computational problems and individual tasks can be offloaded to other devices on the network. But in doing so, the middleware needs to account for limited network throughput and imperfect transmission reliability, data placement issues, the target's device architecture, Quality-of-Service requirements of individual tasks and more while also maintaining context-awareness to achieve optimal resource utilization. Our research implementations include e.g. the Tasklet Trading System and Wasimoff.

Here, we are dedicated to projects in the area of assisting systems involving ubiquitous systems, especially technologies in the areas Internet of Things (IoT) and self-adaptation. These projects are concerned with improving accessibility for people with special physical and mental requirements on the one hand and with maintaining health in general on the other hand. Sensors are often used to measure either motor movements (accelerometer, gyroscope) or physiological data (skin conductivity, pulse, heart rate variability, eye movements). Examples of such projects are self-adaptive fall detection systems and a crowdsourced accessible street map for navigating wheelchair users through cities.

Within the Smart City context we work on projects that focus distributed systems and citizen science. We aim to harness the collective power of sensor networks, technology infrastructure, and active citizen participation to create more intelligent and responsive urban environments. Projects involve developing and deploying sensors to gather diverse data, ranging from environmental factors to traffic patterns. Distributed data structures are then utilized to efficiently process and manage collected data, ensuring real-time insights and enabling various services. Citizen science engages residents in data collection and validation, fostering a sense of ownership and community involvement. This symbiotic approach not only enhances the accuracy and granularity of urban data but also empowers citizens to contribute to the development of their cities, leading to more inclusive and sustainable urban development paradigms.

Board and computer games are a popular leisure activity for all age groups. While these games are usually designed for indoor spaces, they are intended to be brought outdoors in this project, where players take on the roles of the characters and the world becomes the game board. This allows established concepts to be expanded in such a way that they create incentives to go outside and engage in a playful context. In doing so, the possibilities of modern technologies, particularly mobile devices, are leveraged. These technologies are capable of capturing contextual information such as location, device orientation, and speed, and responding accordingly. At the same time, they can provide an engaging user interface for players and intuitively represent various game logics. An example of such games is Scotland Yard. In addition to a technical documentation of the implementation, a short film will be created to explain the game principle and invite people to try out the games.

Energy consumption poses mayor challenges, consuming around 12% of global power. Data centers in Germany alone used 16 billion kWh in 2020. Similarly, households producing surplus solar power face compensation issues due to energy grid limitations. The Follow-The-Sun-Computing project proposes a decentralized solution, distributing computational tasks based on surplus energy, creating a market for computing power exchange. This research identifies and quantifies the benefits of utilizing locally generated energy for distributed computing.

The rise in production of cost-efficient unmanned aerial vehicles ("UAVs" or "drones") opens up new possibilities for process optimizations in industrial settings. The InteGreatDrones project focuses on the use of drone swarms for data collection in dynamic inland terminals, addressing challenges of changing cargo types and stakeholder interactions. The project's goal is to employ autonomous drones to capture terminal activities, ensuring privacy through local data processing and selective sharing. A central middleware connects drones and existing systems, supporting terminal-specific applications like route optimization and predictive maintenance. This project seeks to provide real-time data for informed operational and strategic decisions.