RI: Medium: Collaborative Research: Decision-Making on Uncertain Spatial-Temporal Fields: Modeling, Planning and Control with Applications to Adaptive Sampling

RI：中：协作研究：不确定时空场的决策：建模、规划和控制及其在自适应采样中的应用

• 批准号: 1302360
• 负责人: Marin Kobilarov
• 金额: $ 19.96万
• 依托单位: Johns Hopkins University
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2013
• 资助国家: 美国
• 起止时间: 2013-06-01 至 2018-05-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1302360&HistoricalAwards=false
• 关键词: RI; Medium; Collaborative; Research; Decision

Inland bodies of freshwater are a resource that is critical for the Nation's health and safety. This project is developing a new spatio-temporal field representation suitable for modeling, planning and control under uncertainty in order to improve monitoring of such water systems. The project's focus is on a reconfigurable aquatic sensor-actuator network designed to capture data from coupled physical, chemical, and biological processes that occur across space and time-scales. The key advantages of this sensor-actuator network in its application to this domain include synoptic volume coverage, adaptive sampling, flexible control and robustness to component failure. The research objective is to build models of dynamic processes for which high resolution sampling is necessary at special locations. Toward this end, this project is contributing new methods, data-structures, algorithms, and implementations validated by field testing a heterogeneous system consisting of stationary and mobile (robotic) underwater node. This project provides unique interdisciplinary opportunities for education of both graduate and undergraduate students via new course work that blends projects and research topics directly into courses and newly developed seminars. It provides a multi-disciplinary experience for students while developing their engineering skills. Relevant components of computer science, computer engineering, and mechanical engineering are integrated together by using the project's aquatic platform and experimental scenarios as a focal point. The project advances the state-of-the-art for such systems because it integrates low-level dynamic processes with high-level planning and distributed optimization. The research represents a change in the scale of robotic aquatic sampling away from immense bodies ofwater in oceanographic research, toward bodies of water that have a more immediate affect on our well-being as they are sources and stores of drinking water. The impact of datasets which lead to better understanding of managed and natural inlets, differing topography including dam walls and man-made structures, regions of turbulence, and seasonal algal growth are immense.

淡水的内陆团体是一种对国家健康和安全至关重要的资源。 该项目正在开发一种新的时空场表示，适合在不确定性下建模，计划和控制，以改善对此类水系统的监测。该项目的重点是可重构的水生传感器传导器网络，旨在捕获跨空间和时间尺度发生的物理，化学和生物学过程的数据。 该传感器传动器网络在该域应用程序中的关键优势包括天气体积覆盖，自适应采样，灵活的控制和组件故障的鲁棒性。 研究目标是建立动态过程模型，在特殊位置需要高分辨率采样。为此，该项目正在贡献新的方法，数据结构，算法和实现通过现场测试验证的实现，该系统由固定和移动（机器人）水下节点组成的异质系统验证。该项目通过新的课程工作为研究生和本科生提供独特的跨学科机会，这些课程将项目和研究主题直接融入课程和新开发的研讨会。它在发展工程技能的同时为学生提供了多学科体验。计算机科学，计算机工程和机械工程的相关组件通过使用项目的水生平台和实验场景作为焦点来集成在一起。该项目可以为此类系统的最新设备推进，因为它将低级动态流程与高级计划和分布式优化集成在一起。这项研究代表了机器人水生的规模的变化，从海洋研究中的巨大物体变成了水域，这些水体对我们的福祉产生了更直接的影响，因为它们是饮用水的来源和存储。数据集的影响会导致对托管和自然入口的更好理解，包括大坝墙和人造结构，湍流区域以及季节性藻类生长的不同地形。