A dynamically adaptive system (DAS) monitors itself and its execution environment to evaluate requirements satisfaction at run time. Unanticipated environmental conditions may produce sensory inputs that alter the self-assessment capabilities of a DAS in unpredictable and undesirable ways. Moreover, it is impossible for a human to know or enumerate all possible combinations of system and environmental conditions that a DAS may encounter throughout its lifetime. This paper introduces Loki, an approach for automatically discovering combinations of environmental conditions that produce requirements violations and latent behaviors in a DAS. By anticipating adverse environmental conditions that might arise at run time, Loki facilitates the identification of goals with inadequate obstacle mitigations or insufficient constraints to prevent such unwanted behaviors. We apply Loki to an autonomous vehicle system and describe several undesirable behaviors discovered.

This paper describes a study of the evolution of robust communication, specifically the distribution of data among individuals in a population, using digital evolution. In digital evolution, a population of self-replicating computer programs exists in a user-defined computational environment and is subject to instruction-level mutations and natural selection. To encourage the evolution of this cooperative behavior, we make use of "digital germlines," a form of group-level selection similar to multicellularity in biology. The results of experiments using the Avida platform for digital evolution demonstrate that populations of digital organisms are capable of evolving to distribute data in a network, and that through the application of different selective pressures, these digital organisms can overcome communication obstacles such as message loss, limited bandwidth, and node failure.

It is becoming increasingly important to be able to adapt an application's behavior at run time in response to changing requirements and environmental conditions. Adaptive programs are typically difficult to specify, design, and verify. A variety of conditions may trigger an adaptation, each of which may involve different types of adaptation mechanisms. In many cases, adaptive systems are concurrent, thus further exacerbating the complexity. Furthermore, it is important that adaptations do not put the system into an inconsistent state during or after adaptation. This paper presents an iterative approach to modeling and analyzing UML behavioral design models of adaptive systems, where the UML state diagrams are automatically translated into Promela code for analysis with the Spin model checker. The adaptive models are analyzed for adherence to both system invariants and properties that should hold during adaptation. We demonstrate this approach on applications for the mobile computing domain where we verify the design models against formally-specified properties.