CAREER: Resource-adaptive distributed estimation for teams of micro aerial vehicles

提出的工作集中在为微型飞行器团队(MAV)开发资源自适应的分布式估计算法。在为MAV团队设计估计器时面临的几个挑战中，最重要的一个是MAV的严格资源限制。虽然在任何工程系统中，资源都不可避免地是有限的，但成本、功率和重量的考虑使得对小牛团队的限制特别严格。在这项工作中，我们将为估计和推理算法的设计以及MAV平台本身的设计开发一个严格的、基于优化的框架。我们的方法将产生分布式估计器，能够优化分配MAV团队的传感、处理、通信和能源资源。将开发的方法将导致有效的设计工具，并将允许系统地研究MAV资源可用性和估计性能之间的权衡曲线。由此产生的算法和理论结果将极大地提高MAV团队的能力，从科学探索到搜救行动。反过来，这些系统将产生直接影响我们生活的好处，从促进我们对科学的理解到在灾难现场拯救人类。此外，拟议的研究计划将为加州大学河滨分校不同学生群体的研究生和本科生创造机会，进行有意义的研究。本科生将被招募从事新的MAV设计工作，预计这种参与将增加他们攻读研究生教育的可能性。此外，作为综合推广计划的一部分，我们将利用拟议研究的性质(飞行机器人激发年轻人的想象力)来激励和招募未被充分代表的少数族裔学生进入科学和工程专业。这些努力将有助于缩小代表不足群体的教育程度差距。

The proposed work focuses on developing resource-adaptive distributed estimation algorithms for teams of micro aerial vehicles (MAVs). Among the several challenges one faces when designing estimators for MAV teams, the most important one is the stringent resource limitations of MAVs. While in any engineered system the resources are inevitably finite, cost, power, and weight considerations make the limitations particularly strict for teams of small MAVs. In this work, we will develop a rigorous, optimization-based framework for the design of estimation and inference algorithms, as well as for the design of the MAV platforms themselves. Our approach will yield distributed estimators capable of optimally allocating the sensing, processing, communication, and energy resources of an MAV team. The methods to be developed will lead to efficient design tools, and will permit a systematic study of the tradeoff curves between MAV resource availability and estimation performance.The algorithms and theoretical results that will result from this effort will dramatically increase the capabilities of MAV teams, in domains ranging from scientific exploration to search-and-rescue operations. In turn, these systems will yield benefits that will directly impact our lives, from advancing our state of scientific understanding to saving humans in disaster sites. Additionally, the proposed research plan will create opportunities for both graduate and undergraduate students from UC Riverside's diverse student body to conduct meaningful research. Undergraduate students will be recruited to work on new MAV designs, and it is anticipated that such an involvement will increase the likelihood of them pursuing a graduate education. Moreover, as part of an integrated outreach program, we will leverage the nature of the proposed research (flying robots capture the imagination of young minds) to inspire and recruit underrepresented minority students to science and engineering. These efforts will aid in closing the educational attainment gap for underrepresented groups.