Collaborative Research: Design, Flight Control, and Autonomous Navigation of Bioinspired Morphing Micro Aerial Vehicles for Operation in Confined Spaces

该项目将通过制造可以在极端狭窄空间内运行的蝙蝠无人机，促进空中机器人科学的进步，促进国家的繁荣和福祉。对城市下水道进行快速、连续的环境监测的需求日益增加。这些密闭空间占据了城市基础设施的很大一部分。然而，今天的传统机器人，包括地面和空中系统，无法在绝大多数下水道中运行。下水道网络呈现出一个名副其实的由管道、室和公用设施孔组成的迷宫，这对机器人的运动控制和导航构成了巨大的挑战。因此，人工操作员仍然会检查这些危险空间。除了给人类带来许多风险之外，这些环境中的载人操作成本高且速度慢。这项拨款的研究将有助于设计空中的、受蝙蝠启发的机器人，通过模仿蝙蝠在洞穴中的空中运动原理，这些机器人可以在下水道中工作。这项研究的成果将极大地造福于社会。特别是在大流行期间，使用这些蝙蝠启发的无人机对城市老化的下水道系统进行连续和自动监测，可以在拯救人类生命方面发挥至关重要的作用。例如，sars冠状病毒可以在废水中存在数天，2019年在下水道中及早发现，本可以加强应对当前大流行的准备工作。该项目培养具有跨学科技能的新一代科学家、工程师和技术人员，为未来的专业人士提供垂直整合的、使用启发式的体验式学习活动。与地面机器人相比，如今的旋翼无人机是检查和监控城市基础设施的更好解决方案，因为它们具有可扩展性、廉价、易于部署和快速移动性。然而，这些系统不能在狭窄的区域内飞行，比如小截面的隧道，因为它们依赖于强大而连续的气流。到目前为止，这些无人机的密闭空间应用仅包括在建筑物内部或非常宽敞的密闭环境中飞行，这与空中机器人在下水道中的应用无法相比。这项研究将允许无人机在下水道等狭窄空间内完成完全自主飞行。研究团队将(i)使用集成的机械智能和控制框架来设计具有动态多用途机身构象和显著计算复杂性的无人机，以防止产生强大的空气射流；（ii）设计一个集成的基于模型和数据驱动的飞行控制框架，不仅可以捕获模型的不确定性，还可以捕获由于在隧道中飞行而产生的环境空气动力学效应。（iii）开发一种新的导航框架，该框架依赖于从下水道示意图中提取的高级抽象指导，从而在未映射的环境中实现稳健的导航。该项目由跨部门机器人基础研究项目支持，由工程（ENG）和计算机与信息科学与工程（CISE）联合管理和资助。该奖项反映了美国国家科学基金会的法定使命，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 project will promote the progress of science in aerial robotics and advance the national prosperity and welfare, by creating bat-inspired drones that can operate in extremely confined spaces. There is an increased demand for fast, continuous environmental surveillance in city sewers. These confined spaces occupy a large portion of a city's infrastructure. However, today's conventional robots, including ground and aerial systems, cannot operate in a vast majority of sewers. Sewer networks present a veritable maze of pipes, chambers, and utility holes that pose tremendous challenges for robot locomotion control and navigation. Therefore, human operators still inspect these dangerous spaces. In addition to carrying many risks for humans, manned operations in these environments are costly and slow. This grant’s research will contribute to the design of aerial, bat-inspired robots that can operate in sewers, by mimicking bat aerial locomotion principles in caves. The results from this research will greatly benefit the society. Especially during a pandemic, the continuous and automated monitoring of a city’s aging sewer systems using these bat-inspired drones can play a vital role in saving human lives. For instance, SARS-CoVs can be present in wastewater for several days, and its early detection in sewers in 2019 could have allowed for increased preparedness in combating the current pandemic. This project trains new generation scientists, engineers, and technologists with interdisciplinary skills, providing future professionals with vertically integrated, use-inspired experiential learning activities.Today's rotary-wing drones are better solutions for inspection and monitoring city infrastructure than ground robots because they are scalable, inexpensive, easy to deploy, and possess fast mobility. However, these systems cannot fly inside tight areas such as tunnels with small cross-sections because they rely on powerful and continuous air jets. So far, the confined space applications of these drones include only flying inside buildings or very spacious confined environments, which are not comparable to the application of aerial robots in sewers. This research will allow drones to complete fully autonomous flights inside tight spaces such as sewer galleries. The research team will (i) use an integrated mechanical intelligence and control framework to design drones with dynamically versatile body conformations and significant computational complexity to prevent creation of powerful air jets, (ii) design an integrated model-based and data-driven flight control framework that captures not only the model uncertainty but also the environmental aerodynamic effects that arise due to flight in tunnels, and (iii) develop a novel navigation framework that relies on high-level abstract guidance extracted from sewer schematic diagrams, enabling robust navigation in unmapped environments.This project is supported by the cross-directorate Foundational Research in Robotics program, jointly managed and funded by the Directorates for Engineering (ENG) and Computer and Information Science and Engineering (CISE).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.