Digital worlds offer an excellent opportunity for dynamic storytelling. However, adding inhabitants to a digital world that can adjust to their environment and express interesting unscripted behavior is challenging. This thesis proposes a method to combine short-term planning with long-term optimization to spark emergent behavior and maximize agents’ fitness to the environment. We show that our approach can reliably demonstrate emergent ...

Natural hazards are a driver of ecological dynamics as they alter individual species, community structure and entire ecosystems. Their ecological impact can be highly variable over space and time: perceived by some species as disastrous but vital for the survival of others. The ability of a system to cope with such events is called ecological resilience and depends on biotic and abiotic factors, as well as the disturbance legacies. With ...

The development of heterogeneous CPU-GPU systems for modern data centers in recent years increased the computational power massively. Multi-Instance GPUs (MIGs) allow to split a single physical GPU into multiple independent instances. Thus, more than one user can get exclusive acces to the same physical GPU. However, there are many different configurations for a single MIG-capable GPU that need to be controlled by a central instance. ...

With increasing intermittent distributed power generation, emergency situations in the power system are more likely to occur. Current emergency operation in power systems is mostly based on manual remedial actions. The growing number of distributed generation devices add complexity to emergency operation and make decisions on remedial actions harder. In this thesis, a feedback optimization (FO) controller for emergency power system operation ...