Partially Observable Games for Secure Autonomy

• DOI: 10.48550/arxiv.2002.01969
• Publication Date: 2020-01-01
• Authors: Mohamadreza Ahmadi; Arun Viswanathan; Michel D. Ingham; Kymie Tan; Aaron D. Ames
• Keywords: Testbed; Perfect information; Formalism (music); CONTEST; Decision maker

Technology development efforts in autonomy and cyber-defense have been evolving independently of each other, over the past decade. In this paper, we report our ongoing effort to integrate these two presently distinct areas into a single framework. To end, propose two-player partially observable stochastic game formalism capture both high-level autonomous mission planning under uncertainty adversarial decision making subject imperfect information. We show that synthesizing sub-optimal strategies for such games is possible finite-memory assumptions maker cyber-adversary. then describe an experimental testbed evaluate efficacy proposed