S&AS: INT: COLLAB: Goal-driven Marine Autonomy with Application to Fisheries Science and Management

S

• 批准号: 1848945
• 负责人: Xiaobo Tan
• 金额: $ 22.5万
• 依托单位: Michigan State University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-02-15 至 2022-01-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1848945&HistoricalAwards=false
• 关键词: S& INT; COLLAB; Goal; driven

Marine robots can be used to accurately map and track marine life, leading to a better interpretation of variability and migration patterns that are important to fisheries managers in marine protected areas (MPAs). Principles of engineering and oceanography can be used to maximize the impact of a network of marine robots, but because fisheries managers, oceanographers, and roboticists have different perspectives and knowledge bases, it can be difficult to take advantage of cutting-edge research in each field without significant effort to translate among the groups. Researchers will develop a computational interface that translates human-specified missions of fisheries managers into multi-level planning for a fleet of marine robots to monitor fish populations in a dynamic coastal ocean environment. The system will be designed with input from fisheries managers through a series of workshops, and will be field-tested at Gray's Reef National Marine Sanctuary, a federally-managed MPA off the coast of Georgia. The research will lead to more accurate and effective ways to monitor fish populations in MPAs, as well as breakthroughs in key areas of artificial intelligence and autonomous systems. Many of the results will be applicable to other smart and autonomous systems in challenging environments. In addition, the project will train graduate students and broaden undergraduate education in Science, Technology, Engineering, and Mathematics (STEM), and offer a number of outreach activities, including working with the University of Georgia Marine Extension service to develop a summer camp for middle and high school students.The project is focused on developing an intelligent physical system (IPS) that consists of a heterogeneous fleet of marine robots, cooperatively tracking fish movement and surveying the habitat with minimum request for human intervention. The IPS will translate the human-specified missions of fisheries managers into goal-driven task designs for each robot, and automatically generate executable plans for the networked mobile sensing agents. The system will autonomously and persistently collect in-situ measurements and acoustic detections of fish species while maintaining multi-scale data streams and constructing multiple spatial-temporal maps reflecting the conditions of the ecosystem. The research aims to discover the hotspots (e.g., spatial locations with sustained congregations of fish), as well as illuminate more information about how and when fish move among these hotspots. This goal is quite challenging due to a number of gaps between project needs and the state-of-art autonomy research. Researchers will address the challenges through new developments that accomplish three main tasks: (1) Developing the goal-driven marine autonomy for fish habitat survey, (2) realizing the goal-driven autonomy on physical systems, and (3) evaluating the developed framework through real-life field work, experiments, and data analysis.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

海洋机器人可用于准确绘制和跟踪海洋生物，从而更好地解释变化和迁徙模式，这对于海洋保护区（MPA）的渔业管理者非常重要。 工程和海洋学原理可用于最大限度地发挥海洋机器人网络的影响，但由于渔业管理者、海洋学家和机器人专家拥有不同的观点和知识基础，因此很难利用每个领域的尖端研究无需付出巨大努力在各组之间进行翻译。研究人员将开发一个计算接口，将渔业管理者的人类指定任务转化为海洋机器人舰队的多层次规划，以监测动态沿海海洋环境中的鱼类种群。该系统将根据渔业管理人员通过一系列研讨会提供的意见进行设计，并将在格雷礁国家海洋保护区（乔治亚州海岸附近的一个联邦管理的海洋保护区）进行现场测试。 该研究将带来更准确、更有效的方法来监测海洋保护区的鱼类种群，并在人工智能和自主系统的关键领域取得突破。许多结果将适用于具有挑战性的环境中的其他智能和自主系统。 此外，该项目还将培训研究生并扩大科学、技术、工程和数学 (STEM) 方面的本科教育，并提供一系列外展活动，包括与佐治亚大学海洋推广服务部门合作，为学生举办夏令营该项目的重点是开发一种智能物理系统（IPS），该系统由异构海洋机器人组成，能够合作跟踪鱼类运动并调查栖息地，而对人类干预的要求最低。 IPS 将把渔业管理人员指定的任务转化为每个机器人的目标驱动任务设计，并自动为联网移动传感代理生成可执行计划。该系统将自主、持续地收集鱼类物种的原位测量和声学检测，同时维护多尺度数据流并构建反映生态系统状况的多个时空地图。该研究旨在发现热点（例如，鱼类持续聚集的空间位置），并阐明有关鱼类如何以及何时在这些热点之间移动的更多信息。由于项目需求与最先进的自主研究之间存在许多差距，这一目标相当具有挑战性。研究人员将通过新的发展来应对挑战，完成三个主要任务：（1）开发鱼类栖息地调查的目标驱动的海洋自主，（2）实现物理系统的目标驱动的自主，以及（3）评估所开发的框架通过现实生活中的实地工作、实验和数据分析。该奖项反映了 NSF 的法定使命，并通过使用基金会的智力优点和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

Time-difference-of-arrival (TDOA)-based distributed target localization by a robotic network

机器人网络基于到达时间差 (TDOA) 的分布式目标定位

• DOI: 10.1109/tcns.2020.2979864
• 发表时间: 2020
• 期刊: IEEE Transactions on Control of Network Systems
• 影响因子: 4.2
• 作者: Ennasr, Osama N.;Tan, Xiaobo
• 通讯作者: Tan, Xiaobo

Backstepping Control of Gliding Robotic Fish for Trajectory Tracking in 3D Space

• DOI: 10.23919/acc45564.2020.9147628
• 发表时间: 2020-07
• 期刊: 2020 American Control Conference (ACC)
• 作者: Demetris Coleman;Xiaobo Tan

Randomized Sensor Selection for Nonlinear Systems With Application to Target Localization

• DOI: 10.1109/lra.2019.2928208
• 发表时间: 2019-07
• 期刊: IEEE Robotics and Automation Letters
• 影响因子: 5.2
• 作者: S. Bopardikar;Osama Ennasr;Xiaobo Tan

Observability-aware Target Tracking with Range Only Measurement

通过仅范围测量进行可观测性目标跟踪

• DOI: 10.23919/acc50511.2021.9483280
• 发表时间: 2021
• 期刊: Proceedings of 2021 American Control Conference
• 作者: Coleman, Demetris;Bopardikar, Shaunak D.;Tan, Xiaobo
• 通讯作者: Tan, Xiaobo

Adaptive Parameter Estimation of a Steerable Drifter

可操纵漂流器的自适应参数估计

• DOI: 10.1016/j.ifacol.2021.11.161
• 发表时间: 2021
• 期刊: IFAC-PapersOnLine
• 作者: Gaskell, Eric;Tan, Xiaobo
• 通讯作者: Tan, Xiaobo