EAGER: Game and Teaming Strategies for Networked Systems

将展开理论和计算调查，以解决与分析、控制和设计网络物理系统相关的基本问题，在存在另一组对手的情况下形成一个自主合作的团队。本工作的目标是将先前为特定应用场景开发的理论的范围扩展到更广泛的动态应用集，在这些应用中，博弈论(和控制)可能会发挥作用。这种潜在的应用范围非常广泛，包括机器人、经济建模、能源分配和生物医学。至于这项探索性工作的技术贡献，人们可以从几个不同的角度期待结果。传统博弈论的第一个出发点，即研究动态过程，而不是研究均衡，已经提到了。第二个出发点是，虽然最优控制和系统理论在过去确实利用了博弈论技术，但这本文献在很大程度上限于线性过程。目前的研究将更一般地包括非线性系统，以及混合(即混合连续、离散)和异步系统。因此，在以应用为指导的同时，开发一般理论将是拟议研究的预期结果。第三个出发点是，虽然早期的理论侧重于获得解析解，但目前的研究结果(因为它涉及到非线性过程)本质上必然是计算的或算法的。拟议的理论/计算工作的潜在社会影响范围从经济学、生物医学到国防应用。由于私人投资机构与应用领域和行业密切联系，将创新方法和工具转移到应用程序的潜力是巨大的。PIS还计划对研究生和本科生进行相关主题的教育。

Theoretical and computational investigations will be launched to address fundamental issues associated with analysis, control, and design of networked physical systems forming an autonomously cooperative team in the presence of another team of adversaries. The goal of the present work is to expand the scope of the theory that was developed earlier for the specific applications scenarios to wider set of dynamic applications where game theory (and control) potentially has a role to play. The range of such potential applications is are very broad including robotics, modeling of economies, energy distribution and biomedicine.As for the technical contributions of this exploratory work one can expect results from several different perspectives. The first point of departure from traditional game theory namely that of the study of dynamic processes rather than the study of equilibria has been already mentioned. Second point of departure is that while optimal control and system theory had indeed made use of game theoretic techniques in the past, this body of literature was largely limited to linear processes. The present investigation will be far more general to include nonlinear, as well as hybrid (i.e., mixed continuous, discrete), and asynchronous systems as well. Thus, while guided by applications, developing a general theory would be expected outcome of the proposed research. A third point of departure is while earlier theories focused on obtaining analytical solutions, the results from present investigation (since it involves nonlinear processes) will of necessity be computational or algorithmic in nature.The potential societal implications of the proposed theoretical/computational work is great ranging from economics, biomedicine to defense applications. Since the PIs are in close contact with application domains and the industry the potential for transfer of innovative methods and tools to applications is tremendous. The PIs also plan to educate graduate students as well as undergraduate students on related topics.