NRI: INT: COLLAB: Raining Drones: Mid-Air Release & Recovery of Atmospheric Sensing Systems

了解和预报天气对几乎每个行业都是至关重要的，随着天气模式从历史常态不断变化，这一点更加重要。无人驾驶航空系统(UAS)能够在大气低公里处收集有价值的数据，而传统的大气传感系统无法对其进行采样。然而，一架无人机一次只能从大气层的一小部分收集信息。该项目将开发系统和技术，使UAS能够携带、部署和回收较小的UAS和基于降落伞的天气传感器，以创建跨越大气气团边界的快照。这将导致一个更大的低层大气区域的特征，从而改善对天气的理解和预报。携带、部署系统，特别是恢复系统，对于允许在远程位置进行操作和重复使用昂贵的传感器至关重要。实现这些目标需要新的技术来实现空中机器人之间的丰富的空中互动。这些技术从传感和规划到控制和运行时验证，所有这些技术都在一起工作，以有效、高效和安全地部署这样的空中机器人团队。虽然这项工作是在大气科学的背景下开发的，但这些技术广泛适用于分布式和协作机器人的许多问题。该项目将贡献技术和系统，使执行空中捕获和释放其他系统的UAS团队的开发和部署成为可能。更具体地说，这项工作的预期结果包括：1)用于UAS拦截具有随机全六自由度运动的空中目标的传感、规划和控制方法，2)用于不同类别空中机器人之间的匹配机动以执行空中对接的基本要素，3)用于UAS团队的快速空中部署-捕获-重新部署周期的策略，以使得能够在大地理尺度上进行观察，4)运行时推理和执行协调分布式、不同种类机器人系统之间的交互的协议，和5)改进的大气传感和监测能力，将增强对大气现象的理解和预测。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

Understanding and forecasting weather is critical to nearly every industry and is even more important as weather patterns continue to change from their historical norms. Unmanned Aerial Systems (UASs) are capable of collecting valuable data in the lower kilometer of the atmosphere, which is under-sampled by traditional atmospheric sensing systems. However, a single UAS can only collect information from a small slice of the atmosphere at a time. This project will develop the systems and techniques to enable a UASs to carry, deploy, and recover smaller UASs and parachute-based weather sensors to create snapshots across atmospheric airmass boundaries. This will lead to the characterization of a much larger region of the lower atmosphere, consequently improving the understanding and forecasting of the weather. Carrying, deploying, and particularly recovering the systems, are critical to allow operations in remote locations and reuse of expensive sensors. Achieving these objectives requires new techniques to enable rich mid-air interactions between aerial robots. These techniques span from sensing and planning to control and run-time verification, all working together for the effective, efficient, and safe deployment of such teams of aerial robots. While this work is developed in the context of atmospheric sciences, the techniques are broadly applicable to many problems in distributed and collaborative robotics.This project will contribute techniques and systems to enable the development and deployment of teams of UASs that perform mid-air capture and release of other systems. More specifically, the expected outcomes of this work include: 1) Sensing, planning and control methodologies for UASs intercepting airborne targets that have stochastic full six degree of freedom motion, 2) Foundational elements for matched maneuvers between heterogeneous classes of aerial robots to perform aerial docking, 3) Strategies for rapid aerial deployment-capture-redeployment cycles for teams of UASs to enable observations over large geographic scales, 4) Run-time inference and enforcement of protocols orchestrating the interactions between distributed, heterogeneous robotic systems, and 5) Improved atmospheric sensing and monitoring capabilities that will enhance understanding and forecasting of atmospheric phenomena.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.