Collaborative Research: NRI: Dispersed Autonomy for Marsupial Aerial Robot Teams

该项目将进行机器人技术的创新整合，以创建有袋动物空中机器人团队（MARTs），由小型无人驾驶飞机系统（即无人机）携带小型飞机进行大气观测。利用mart收集的数据，有效的先进预警系统可以大大减少恶劣天气造成的生命损失和损害。在对人类来说太危险的地方运行的空中机器人系统或昂贵的单飞行器系统将需要对详细的模型和大量的数据进行推理。这个项目将创建和部署一个机器人团队，就像一个自主的空中气象学家，进行在线有针对性的预报。MARTs实现了国家机器人计划的目标，即促进机器人的整合以造福人类。该项目将开发一种自主的空中气象学家，它由连接到分散用户和计算资源的空中机器人团队组成。通过在有袋类空中机器人团队中集成、部署和评估新的基本算法，推进当前实践，体现了机器人技术创新整合的主题。集成机器人技术包括：i.)空射伪拉格朗日漂移器；Ii)自主目标观测的集合灵敏度分析；iii.)求解在线连续域部分可观察马尔可夫决策过程（pomdp）的算法；不确定流的并行路径规划；v)有袋动物空中机器人团队的分散自治架构。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 conduct innovative integration of robotics technologies to create Marsupial Aerial Robot Teams (MARTs) whereby small uninhabited aircraft systems (i.e. drones) carry smaller aircraft to provide observations of the atmosphere. Effective advanced warning systems enabled by the data collected by MARTs could drastically reduce the loss of life and damage caused by severe weather. Aerial robotic systems operating in places too dangerous for humans or expensive single-vehicle systems will need to reason over detailed models and large amounts of data. This project will create and deploy a team of robots that will be like an autonomous airborne meteorologist performing online targeted forecasting. MARTs achieve the National Robotics Initiative’s goal to promote the integration of robots to the benefit of humans. This project will develop an autonomous airborne meteorologist that consists of teams of aerial robots connected to dispersed users and computing resources. The effort embodies the theme of innovative integration of robotic technologies by advancing current practices through new fundamental algorithms integrated, deployed, and evaluated on marsupial aerial robot teams. Integrated robotic technologies include: i.) air-launched pseudo-Lagrangian drifters; ii.) ensemble sensitivity analysis for autonomous targeted observation; iii.) algorithms that solve online continuous-domain partially-observable Markov decision processes (POMDPs); iv.) parallelized path planning in uncertain flows; and v.) a dispersed autonomy architecture for marsupial aerial robot teams.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.