Crowd-Sensing Meets Situation Awareness: A Research Roadmap for Crisis Management

Abstract

When disaster strikes, human lives may depend upon emergency organizations' rapid establishment of Situation Awareness (SAW) to take the appropriate decisions and actions. Recently, systems emerged, enabling humans to act as crowd sensors contributing valuable crisis information via mobile devices through social media channels. This should allow enhancing situational pictures gained through traditional SAW systems, as employed in control center domains. A common understanding about the necessary functionality of such crowd-sensed SAW systems for crisis management, however, is not yet reached nor is a detailed comparison thereof available up to now. This paper makes a first attempt towards this by a reference architecture incorporating crowd-sensed crisis information into SAW systems. Based on that, a systematic catalog of evaluation criteria is derived and used for an in-depth comparison of nine existing systems, thereby highlighting current capabilities and directions for further research.

Role Playing Next Generation 9-1-1: Sensemaking with Social Media in Public-Safety Answering Points

“…These approaches can involve collecting data produced by smartphone sensors or public data produced by citizens using social media applications such as Twitter and Instagram to, for example, monitor urban mobility patterns [19] or track shifts in public sentiment around major events [29]. The collection and analysis of social media data has received significant attention as a way to support situational awareness among public safety officials managing emergency services [32], with studies examining the use of social media among police officers [22], public information officers [14], and emergency managers [9].…”

Situation awareness in crowdsensing for disease surveillance in crisis situations

“…Mobile4D-SA follows the general architecture of the JDL data fusion model [9,12]. The Sensing level (L0), consisting of report receipt and feature extraction, receives raw reports from mobile devices and tags them with a time and geo-reference coordinate.…”

“…Yet situation awareness in crowdsensing for crisis management remains a largely unexplored area of research. Indeed, in a recent review of crowdsensing systems for crisis management, Salfinger et al [12] concluded that the core situation awareness functions of integrated perception, comprehension, and projection remain unsupported by existing systems.…”

Crowd^SA — towards adaptive and situation-driven crowd-sensing for disaster situation awareness

“…the IF levels these address (as specified in the JDL data fusion model [17]): Sensing data from the observed environment (JDL Level 0), assessing objects from these measurements (JDL Level 1), assessing the overall ongoing situations (JDL Level 2), projecting these situations' development and impact (JDL Level 3), and furthermore, resource management or process refinement (JDL Level 4) and user refinement (JDL Level 5). In our evaluation, we contrasted the following approaches [18]: HADRian [7], ESA [8], Twitris [9], Twitcident [10], SensePlace2 [11], CrisisTracker [12], TweetTracker [13], Toretter [14], and CIACM [15]. Reviewed from the perspective of a comprehensive SAW architecture stretching across all IF levels, it became apparent that current systems often lack means for automated SA (which is, except for HADRian [7], deferred to the human operator), thus do not support JDL levels 2+, and mainly rely on human expertise and interaction.…”