Counter-Misdirection in Behavior-based Multi-robot Teams

Chen, Shengkang; Arkin, Ronald C.

doi:10.1109/isr50024.2021.9419536

Cited by 2 publications

(1 citation statement)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In our study, the defenders are not learning an HLP, instead the defenders use a hardcoded function to guess the goal by observing the invader's position. Thus, this chapter studies deception and not counter-deception as mentioned in [86]. The defenders believe that the invader has approached the true goal in the past t d seconds.…”

Section: Defenders' Higher-level Policymentioning

confidence: 99%

Learning of Deception in Adversarial Games with Hierarchical Multiobjective Reinforcement Learning

Asgharnia Gourabjiri

View full text Add to dashboard Cite

This thesis focuses on modelling deception in competitive differential games. By definition, differential games are games where two agents (or two groups of agents) have conflicting objectives. In other words, maximizing an agent's (or a group of agents') objective will minimize the other agent's (or a group of agents') pay-off. To increase the pay-off of one group, we implement deception in a differential game. On the other hand, the other group learns counter-deception to cope with the adversarials' strategy. To address deception and counter-deception, we implement the game of guarding territory and its derivatives.Our primary attempt consists of modelling deception using a hierarchical structure for pursuit-evasion games. The deception structure is made of two levels: a lower-level policy (LLP) that contains the optimal state-action values to reach a goal and a higher-level policy (HLP) that can change the lower-level policy's goal. While the players try to deceive their opponents using deception, the opponents try to detect deception and overcome the deceitful players. Reinforcement learning is the primary learning strategy in our study. We address the assignment problem in pursuit-evasion games, where multiple agents exist on each side.The assignment problem is solved by proposing a two-level policy system, where the pursuer's policy is calculated via a fuzzy actor-critic learning (FACL) algorithm. Our study is the first study on pursuit-evasion games where the assignment can be changed in the middle of the simulation. In addition, we proposed a hierarchical reinforcement learning scheme to provide broader control over the reward function hyperparameters. In other words, an HLP learns reward functions' hyperparameters.We combined all the proposed two-level structures. As a result, for the first time, we Finally, I would like to thank my family and friends for their infinite support and encouragement during my studies in the past years. iv Contents x 5.6 Pareto fronts for F itness 1 and F itness 2 in (5.5) in bi-objective optimization. The Pareto fronts are for the three invader's HLP under two speed scenarios and three different set of inputs for the invasion of each goal. . . . . . . . . . 5.7 Deceptive case, where the speeds are equal and the invader's HLP uses Input 1 5.8 Deceptive case, where the speeds are equal and the invader's HLP uses Input 2 5.9 Deceptive case, where the speeds are equal and the invader's HLP uses Input 3 5.10 Deceptive case, where the invader is superior and the invader's HLP uses Input 1 5.11 Deceptive case, where the invader is superior and the invader's HLP uses Input 2 5.12 Deceptive case, where the invader is superior and the invader's HLP uses Input 3

show abstract

Section: Defenders' Higher-level Policymentioning

confidence: 99%