CAREER: Efficient Multi-Vehicle Coordination for Distributed Sensing and Estimation

团体协调车辆的使用正在彻底改变在空中、水下和地面环境中进行观察、测量和估计的方式。特别是，多车辆系统特别适合于在偏远地区进行主动监测和协调行动。它们具有现场实时测量的能力，以及动态适应和与环境集成的额外能力，使其在广泛的场景中优于静态传感器网络。将单个车辆的运动、传感、通信和计算能力整合到具有可预测行为的复杂鲁棒系统中的科学仍处于初级阶段。该项目旨在开发新的工具来分析分布式系统及其设计的复杂性，重点关注分布式估计、能量约束操作和车辆动力学等方面。研究计划的重点如下：(i)研究用于分散估计物理过程的空间插值方法，并综合可证明正确的最佳感知协调算法；（ii）考虑到多车辆系统的非平凡动力学的节能协调政策的设计。创新的技术方法依赖于一组有前途的建模和分析结果，最近由PI和合作者开发，基于几何优化，图论，系统和控制理论。更广泛的影响：本研究将对科学领域和救灾行动中多车辆系统的使用产生重大影响；也就是说，在那些实时遥感信息和数据收集很重要的场景中。在与加州大学圣地亚哥分校斯克里普斯海洋学研究所的合作中，PI计划为用于气候和大气研究的多架无人驾驶飞行器实现高效的协作算法。该提案的其他教育活动包括：(i)创建一个自动多车辆系统实验室，该实验室将作为模拟和验证协作算法的工具，并通过设计项目和暑期实习让本科生参与研究。（ii）在加州数学与科学暑期学校（COSMOS）项目中为有才华的高中生提供一门课程，重点是机器人系统及其协调

The use of groups coordinated vehicles is revolutionizing the way in which observation, measurement, and estimation is carried out in aerial, underwater, and ground environments. In particular, multi-vehicle systems are uniquely suited for active monitoring and coordinated action in remote areas. Their ability to take in-situ, real-time measurements, together with the additional capacity for dynamic adaptation and integration with the environment makes them preferable over static sensor networks in a wide range of scenarios. The science of integrating the motion, sensing, communication, and computation capabilities of individual vehicles into complex and robust systems with predictable behavior is still at a primitive stage. This project aims to develop novel tools to analyze the complexity of distributed systems and their design, focusing on aspects such as distributed estimation, energy-constrained operation, and vehicle dynamics.The research plan is articulated along the following thrusts: (i) the investigation of spatial interpolation methods for decentralize estimation of physical processes and the synthesis of provably correct coordination algorithms for optimal sensing; and (ii) the design of energy-efficient coordination policies that take into account the non-trivial dynamics of multi-vehicle systems. The innovative technical approach relies on a set of promising modeling and analysis results, recently developed by the PI and collaborators, based on geometric optimization, graph theory, and systems and control theory.Broader Impacts: This research will have a significant impact on the use of multi-vehicle systems in scientific domains and disaster relief operations; that is, in those scenarios where real-time remotely sensed information and data gathering are important. In collaboration with the Scripps Institution of Oceanography at UCSD, the PI plans to work on the implementation of efficient cooperative algorithms for multiple unmanned aerial vehicles used in climate and atmospheric research.Additional educational activities of this proposal include: (i) The creation of an Autonomous Multi-Vehicle Systems Laboratory, which will serve as a tool to simulate and validate cooperative algorithms, and as a means to involve undergraduates in research via design projects and summer internships. (ii) The offering of a course for talented high-school students in the California Summer School of Mathematics and Science (COSMOS) program, with a focus on robotic systems and their coordination