EAGER SitS: Active Self-Boring Robots that Enable Next Generation Dynamic Underground Wireless Sensing Networks: Fusion of Fast Prototyping, Modeling and Learning

这个早期概念的探索性研究（EAGER）土壤信号（SitS）奖资助了一个高风险/高回报的生物启发研究项目，该项目需要在地面上开发自钻传感器探头。 分布式传感是基于丰富的空间和时间数据实现自然和建筑环境智能评估或管理的必要条件。无线传感网络（WSN）是特别有吸引力的，因为它消除了昂贵和繁琐的有线连接。然而，在恶劣的环境中，如近地表的地下区域，主要是由于土壤是不透明的，异质的，耗散和难以穿透/挖掘的事实，涉及无线传感器网络的应用存在巨大的挑战。这个EAGER项目探索了一种生物启发的平台技术，称为主动自钻孔机器人，它将使下一代动态地下WSN（DUWSN）成为可能。在这样的网络中，传感器节点被集成到机器人中，模仿穴居动物，以最小的土壤干扰和最小的人为干预自主部署。由于它的机动性，每个机器人都能够返回地面进行服务。更重要的是，这些机器人节点将能够移动到地下，并根据需要改变它们的传感位置。通过这种方式，可以建立动态可重构的而不是静态的无线传感网络，以提高空间覆盖和数据的分辨率以及数据传输的可靠性。拟议的研究具有潜在的变革性，因为小型，敏捷的地下机器人可以被部署用于广泛的直接应用。例子包括精准农业、污染物监测和预测以及堤坝、水坝和地基等基础设施的健康监测。该平台技术还可用于监视、侦察和勘探目的。该奖项使研究团队能够积极接触相关领域的研究人员，探索更广泛的合作，并寻求工业合作伙伴，以加快拟议技术的开发和应用。与亚利桑那州立大学的NSF生物医学和生物启发岩土工程研究中心合作，计划开展外展和教育活动，以吸引代表性不足的第一代本科生和高中生。该项目将通过一个专门用于探索土壤/介质相互作用的机器人自钻平台，促进我们对地下运动的理解。它将回答探索性的问题，如：1）什么是机械要求的渗透剂，以实现运动？2)我们如何模仿各种穴居动物采用的自我钻孔机制？3)为了模拟某种钻孔机构，哪种驱动和控制策略是最好的？4）对于某种类型的土壤，如砂、粉土和粘土，哪种自钻机构是最佳的？寻找解决方案将从检查广泛的生物样本开始。一系列的机器人原型的基础上，不同的自钻孔机制和驱动策略将被设计。将建立一个快速原型策略，系统地设计和制造机器人，并将通过分析和数值建模。将使用实验测试台来表征具有受控运动学和土壤条件的机器人的穿透性能。最后，将建立控制输入、机器人运动学、土壤特性和穿透性能之间的关系，并可能借助机器学习。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

This EArly-concept Grant for Exploratory Research (EAGER) Signals in the Soil (SitS) award funds a high risk/high return bio-inspired research project needed for the development of self-boring sensor probes in the ground. Distributed sensing is essential for realizing intelligent assessment or management of the natural and built environment based on rich spatial and temporal data. A wireless sensing network (WSN) is particularly attractive since it eliminates costly and cumbersome wire connections. However, there are tremendous challenges involved in the application of WSN in harsh environments such as the near-surface underground region, mainly due to the fact that soil is opaque, heterogeneous, dissipative and hard to penetrate/excavate. This EAGER project explores a bioinspired platform technology known as active self-boring robots, which will enable next generation dynamic underground WSN (DUWSN). In such a network, the sensor nodes are integrated into a robot, mimicking burrowing animals, that deploys itself autonomously with minimal disturbance to the soil and minimal human intervention. Thanks to its motility, each robot would be able to return to the surface for service purposes. More importantly, these robotic nodes would be able to locomote underground and change their sensing locations as needed. In this way, a dynamic reconfigurable rather than a static wireless sensing network can be established to improve spatial coverage and the resolution of the data as well as the reliability of the data transmission. The proposed research is potentially transformative since small, agile underground robots can be deployed for a wide range of direct applications. Examples include precision agriculture, contaminant monitoring and prediction, and health monitoring for infrastructure such as levees, dams and foundations. This platform technology can also be used for surveillance, reconnaissance and exploration purposes. This award allows the research team to actively reach out to researchers in related fields and explore broader collaborations, as well as to seek industrial partners to expedite the development and application of the proposed technology. Outreach and educational activities are planned to engage underrepresented first-generation undergraduate and high school students, in collaboration with the NSF Engineering Research Center for Biomediated and Bioinspired Geotechnics at Arizona State University.This project will advance our understanding of underground locomotion via a robotic, self-boring platform purpose-built to explore soil/agent interactions. It will answer exploratory questions such as: 1) What are the mechanical requirements for a penetrating agent to achieve motility? 2) How do we mimic the self-boring mechanisms adopted by various burrowing animals? 3) To mimic a certain boring mechanism, which actuating and control strategy is best? and 4) Which self-boring mechanism is optimal for a certain type of soil such as sand, silt and clay? The search of solutions will start by examining a wide array of biological exemplars. An array of robot prototypes based on different self-boring mechanisms and actuating strategies will be designed. A fast prototyping strategy will be established to systematically design and fabricate the robots and will be informed by analytical and numerical modeling. The penetration performance of the robots with controlled kinematics and soil conditions will be characterized using an experimental testbed. And finally, relationships will be established to correlate control input, robot kinematics, soil properties and penetration performance, possibly aided by machine learning.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.