arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2407.10107math.OCcs.SYeess.SY
收藏

双人零和混合博弈

Two-Player Zero-Sum Hybrid Games

Santiago J. Leudo, Ricardo G. Sanfelice

AI总结:

本文构建混合动力约束下的双人零和博弈,给出类HJBI方程充分条件保证解存在,最优策略下无需计算混合系统解即可求值函数,还证明最优反馈律可使目标集渐近稳定,可用于扰动抑制等鲁棒控制场景。

摘要:

本文构建了一类受混合动力方程定义的动态约束下的双人零和博弈。该博弈对应一个极小-极大问题,其代价泛函依赖于博弈双方的行动以及混合系统的对应解,这些解被定义为混合时间的函数,因此可以发生流或跳变。通过合理定义终端集,可将有限时域和无限时域问题统一纳入该框架。本文给出了类Hamilton-Jacobi-Bellman-Isaacs方程形式的充分条件,以保证博弈解的存在性。研究表明,当博弈双方选择最优策略时,无需计算混合系统的解即可求解值函数。在额外条件下,本文证明最优状态反馈律可使目标集合对所得混合闭环系统渐近稳定。这类博弈的应用以鲁棒控制问题的形式呈现,包括扰动抑制和安全问题。

英文摘要:

In this paper, we formulate a two-player zero-sum game under dynamic constraints defined by hybrid dynamical equations. The game consists of a min-max problem involving a cost functional that depends on the actions and resulting solutions to the hybrid system, defined as functions of hybrid time and, hence, can flow or jump. A terminal set conveniently defined allows to recast both finite and infinite horizon problems. We present sufficient conditions given in terms of Hamilton-Jacobi-Bellman-Isaacs-like equations to guarantee to attain a solution to the game. It is shown that when the players select the optimal strategy, the value function can be evaluated without computing solutions to the hybrid system. Under additional conditions, we show that the optimal state-feedback laws render a set of interest asymptotically stable for the resulting hybrid closed-loop system. Applications of these games, presented here as robust control problems, include disturbance rejection and security problems.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑