CPS EAGER: Intelligent Agent Incident Command System Augmentation

智慧城市是复杂的网络物理系统，人口众多，增加了额外的复杂性。仪表和建模是智慧城市的组成部分。然而，不管仪器的无处不在和模型的精度如何，最终，人类和人类团队将对全市范围的运营和管理做出决策，尤其是在危机中。我们认为，当代指挥和控制系统的等级本质可能造成虚拟盲点，在这些盲点中，机会或危险可能对等级来说是不可见的，因为必要的信息在等级中的抽象层次之间移动时被模糊了。该项目将包括与危机管理专家和研究人员合作，开发智能代理，旨在最大限度地减少决策者的认知负荷，利用报告和传感器数据中的语义关系来建议其他看不见的事件，并对地面资产的行为进行排序，以改进因果关系模型，以更好地反映危机和/或事件管理期间的持续发展。该项目解决了从研究发现到商业应用的以下技术差距：a)在代理可以在不增加人类认知工作量的情况下支持和增强人类决策的情况下，展示了信息呈现和透明度的有效性；B)将语义数据和信息集成方面的最新基础研究转移到复杂的灾难场景中；C)开发模型一致性维护工具，用于自动更新各种灾害和/或紧急情况的因果模型。此外，参与该项目的人员，例如研究生，将通过所开发模型的设计、开发和测试获得创新经验。该项目将探索将研究成果转化为其他应用领域的工具，如跳伞救援培训、AFRL灾难响应系统RIPPLE和克拉克县应急管理局。该项目还将开展拓展工作，指导探索实验室Tec^Edge的高中生和本科生参加Edge暑期/年度项目（SATE/YATE）。

Smart Cities are complex cyber-physical systems with large human populations adding additional complexity. Instrumentation and modeling are components of a smart city. Regardless, however, of the ubiquity of instrumentation and precision of models, in the end, humans and human teams will make decisions about citywide operations and management, especially in crisis. We contend that the hierarchical nature of contemporary command and control systems can create virtual blind spots in which opportunities or dangers may be invisible to the hierarchy because the necessary information is obscured as it moves between levels of abstraction in the hierarchy. This project will involve teaming with crisis management experts and researchers to develop intelligent agents designed to minimize cognitive load on decision makers, exploit semantic relationships in reports and sensor data to advise of otherwise invisible occurrences, and sequence the actions of ground-level assets to refine causal relationship models to better reflect ongoing developments during crisis and/or event management. This project addresses the following technology gap(s) as it translates from research discovery toward commercial application - a) demonstration of the effectiveness of information presentation and transparency in situations where agents can support and enhance human decision-making without increasing the cognitive workload of the human; b) transfer state-of-the-art foundational research in semantic data and information integration to the complex disaster scenario; c) development of model consistency maintenance tools for automatic update of causal models of various disaster and/or emergency situations. In addition, personnel involved in this project, e.g., graduate students, will receive innovation experiences through the design, development and testing of the model developed.This project will explore transferability of the research results into tools in other application areas such as Pararescuer training, AFRL disaster response system RIPPLE, and Clark County Emergency Management Agency. This project will also have outreach efforts with mentoring high school and undergraduate students at Discovery Lab, Tec^Edge through the Summer at the Edge/Year at the Edge Programs (SATE/YATE).