CHS: Small: Collaborative Research: Spatio-Temporal Situational Awareness in Large-Scale Disasters Using Low-Cost Unmanned Aerial Vehicles

CHS：小型：合作研究：利用低成本无人机实现大规模灾害中的时空态势感知

• 批准号: 1528137
• 负责人: David Feil-Seifer
• 金额: $ 16.67万
• 依托单位: Board of Regents, NSHE, obo University of Nevada, Reno
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2016
• 资助国家: 美国
• 起止时间: 2016-01-01 至 2018-12-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1528137&HistoricalAwards=false
• 关键词: CHS; Small; Collaborative; Research; Spatio

Providing real-time situational awareness is a critical but immensely challenging component in the management of large-scale disasters such as wildfires, where incident managers need to make timely decisions and allocate resources. Yet gathering accurate real-time information is difficult due to the multi-scale spatio-temporal nature of the event, the overwhelming amount of data that has to be processed in near real-time, the heterogeneous nature of the technological resources involved, and the complex interdependencies and interactions between human and technological entities. Recent advances in various areas of cyber-physical and information systems, including sensing, mapping, communication, computing technologies, and unmanned systems, have provided an unprecedented opportunity to revolutionize the acquisition of situational awareness in a large-scale disaster. The objective of this multi-institutional project is to conduct fundamental research aimed at creating user-centered control and algorithmic tools that integrate real-time sensory data from multi-rotor unmanned aerial vehicles (UAVs) into effective fire-predictor software, which will allow a team of UAVs equipped with electro-optical sensors to generate situational awareness in a large-scale wildfire. The PIs will develop a reliable and affordable UAV system that is portable, safe, and easy to operate by first responders (especially firefighters and forestry officials), and which will enable them to make informed decisions which can optimize resource allocation and thereby save both property and lives. With the imminent inclusion of UAVs in the national airspace, the technologies developed in this research will have potential broad applicability to a number of other civilian applications such as law enforcement and border patrol. Project objectives will be met by pursuing specific aims including the design of user-centered analytical and algorithmic tools for robust and safe motion control of UAVs for gathering information, the development of spatio-temporal situational awareness using real-time data and a fire propagation model, and the development of a command, control and communication (C3) framework for emergency management personnel integration. The PIs will create new algorithms for UAV trajectory generation that resourcefully carry out monitoring in a dynamic environment, along with new cooperative control algorithms for a team of UAVs tasked with dynamic perimeter tracking that are robust to addition and/or deletion of UAVs. They will devise distributed filtering methods coupled with reduced order modeling techniques based on real-time spatio-temporal decomposition, to seamlessly generate situational awareness in a computationally efficient manner. Finally, the PIs will develop simulation and field capacity, while studying human-robot interaction issues pertaining to the wildfire application. The PIs' user-centered design approach is novel, in that it will address not only usability issues related to the developed system, but also issues such as deployment, training, and changes in operational paradigms. To ensure project outcomes meet the needs of the target user community, the PI team will be working in collaboration with the City of Cincinnati's Fire Department and the State of West Virginia's Division of Forestry.

提供实时情境意识是管理大规模灾难（例如野火）的关键但极具挑战性的组成部分，事件经理需要及时做出及时的决策并分配资源。 然而，由于事件的多尺度时空性质，必须在近乎实时的实时处理，涉及的技术资源的异质性质以及人类和技术实体之间的复杂相互依存关系和相互作用。 网络物理和信息系统各个领域的最新进展，包括感应，映射，通信，计算技术和无人驾驶系统，为在大规模灾难中彻底改变了对情境意识的获取提供了前所未有的机会。 The objective of this multi-institutional project is to conduct fundamental research aimed at creating user-centered control and algorithmic tools that integrate real-time sensory data from multi-rotor unmanned aerial vehicles (UAVs) into effective fire-predictor software, which will allow a team of UAVs equipped with electro-optical sensors to generate situational awareness in a large-scale wildfire. PIS将开发一个可靠且负担得起的无人机系统，该系统可容纳急救人员（尤其是消防员和林业官员）的便携式，安全且易于操作，这将使他们能够做出明智的决定，从而可以优化资源分配，从而节省财产和生命。 随着在国家领空中迫在眉睫的无人机，这项研究中开发的技术将对许多其他平民应用具有广泛的适用性，例如执法和边境巡逻队。 将通过追求特定目标来实现项目目标，包括设计以用户为中心的分析和算法工具，用于对无人机的强大和安全运动控制，以收集信息，以实时数据和消防传播模型的发展时空情境意识的发展以及指挥，控制和通信（C3）框架（C3）以应对应急管理人员整合的框架。 PI将为无人机轨迹生成创建新的算法，该算法在动态环境中足智时地进行监视，以及针对具有动态周围跟踪的无人机团队的新合作控制算法，这些算法具有动态周围的跟踪，这些算法可为无人机的增加和/或删除。 他们将设计分布式过滤方法以及基于实时时空分解的减少顺序建模技术，以以计算有效的方式无缝地产生情境意识。 最后，PI将在研究与野火应用有关的人类机器人相互作用问题的同时，发展模拟和现场容量。 PIS的以用户为中心的设计方法是新颖的，因为它不仅可以解决与开发系统有关的可用性问题，还将解决部署，培训和操作范式变化等问题。 为了确保项目成果满足目标用户社区的需求，PI团队将与辛辛那提市消防局和西弗吉尼亚州林业部合作合作。