crowd — Towards Adaptive and Situation-Driven Crowd-Sensing for Disaster Situation Awareness

Disasters pose severe challenges on emergency responders, who need to appropriately interpret the situational picture and take adequate actions in order to save human lives. Whereas Information Fusion (IF) systems have proven their capability of supporting human operators in rapidly gaining Situation Awareness (SAW) in control center domains, disaster management presents novel challenges: Due to the unpredictability, uniqueness and large-scale dimensions of disasters, their situational pictures typically cannot be extensively captured by sensors — a substantial amount of situational information is delivered by human observers. The ubiquitous availability of social media on mobile devices enables humans to act as crowd sensors, as valuable crisis information can be broadcast over social media channels. Although various systems have been proposed which successfully demonstrate that such crowd-sensed information can be exploited for disaster management, current systems mostly lack means for automated reasoning on these information, as well as an integration with structured data obtained from other sensors. Therefore, in the present work we provide a first attempt towards comprehensively integrating social media-based crowd-sensing in SAW systems: We contribute an architecture on an adaptive SAW framework exploiting both, traditionally sensed data as well as unstructured social media content, and present our initial solutions based on real-world case studies.