arXiv subjects

Michael Dorothy

Publications and source records attributed to Michael Dorothy.

At least 19 recordsLinked to original sources

arXiv2026

Optimal Hiding with Partial Information of the Seeker's Route

We consider a hide-and-seek game between a Hider and a Seeker over a finite set of locations. The Hider chooses one location to conceal a stationary treasure, while the Seeker visits the locations sequentially along a route. As the search progresses, the Hider observes a prefix of the Seeker's route. After observing this information, the Hider has the option to relocate the treasure at most once to another unvisited location by paying a switching cost. We study two seeker models. In the first, the Seeker is unaware of the fact that the Hider can relocate. In the second, the Seeker select its route while accounting for the possibility that the Hider observes its path and reallocates. For the restricted case, we define the value-of-information created by the reveal and derive upper bounds in terms of the switching cost using a worst-case evaluation over routes. We also show that seeker awareness reduces the game value, with the difference between the restricted and feedback models bounded by the entry-wise gap between the corresponding payoff matrices. Numerical examples show how this benefit decreases as the switching cost increases and as the reveal occurs later along the route.

eess.SY↗

arXiv2026

Game-Theory-Assisted Reinforcement Learning for Border Defense: Early Termination based on Analytical Solutions

Game theory provides the gold standard for analyzing adversarial engagements, offering strong optimality guarantees. However, these guarantees often become brittle when assumptions such as perfect information are violated. Reinforcement learning (RL), by contrast, is adaptive but can be sample-inefficient in large, complex domains. This paper introduces a hybrid approach that leverages game-theoretic insights to improve RL training efficiency. We study a border defense game with limited perceptual range, where defender performance depends on both search and pursuit strategies, making classical differential game solutions inapplicable. Our method employs the Apollonius Circle (AC) to compute equilibrium in the post-detection phase, enabling early termination of RL episodes without learning pursuit dynamics. This allows RL to concentrate on learning search strategies while guaranteeing optimal continuation after detection. Across single- and multi-defender settings, this early termination method yields 10-20% higher rewards, faster convergence, and more efficient search trajectories. Extensive experiments validate these findings and demonstrate the overall effectiveness of our approach.

cs.LG↗

arXiv2026

C-IDS: Solving Contextual POMDP via Information-Directed Objective

We study the policy synthesis problem in contextual partially observable Markov decision processes (CPOMDPs), where the environment is governed by an unknown latent context that induces distinct POMDP dynamics. Our goal is to design a policy that simultaneously maximizes cumulative return and actively reduces uncertainty about the underlying context. We introduce an information-directed objective that augments reward maximization with mutual information between the latent context and the agent's observations. We develop the C-IDS algorithm to synthesize policies that maximize the information-directed objective. We show that the objective can be interpreted as a Lagrangian relaxation of the linear information ratio and prove that the temperature parameter is an upper bound on the information ratio. Based on this characterization, we establish a sublinear Bayesian regret bound over K episodes. We evaluate our approach on a continuous Light-Dark environment and show that it consistently outperforms standard POMDP solvers that treat the unknown context as a latent state variable, achieving faster context identification and higher returns.

eess.SY↗

arXiv2025

IMAS$^2$: Joint Agent Selection and Information-Theoretic Coordinated Perception In Dec-POMDPs

We study the problem of jointly selecting sensing agents and synthesizing decentralized active perception policies for the chosen subset of agents within a Decentralized Partially Observable Markov Decision Process (Dec-POMDP) framework. Our approach employs a two-layer optimization structure. In the inner layer, we introduce information-theoretic metrics, defined by the mutual information between the unknown trajectories or some hidden property in the environment and the collective partial observations in the multi-agent system, as a unified objective for active perception problems. We employ various optimization methods to obtain optimal sensor policies that maximize mutual information for distinct active perception tasks. In the outer layer, we prove that under certain conditions, the information-theoretic objectives are monotone and submodular with respect to the subset of observations collected from multiple agents. We then exploit this property to design an IMAS$^2$ (Information-theoretic Multi-Agent Selection and Sensing) algorithm for joint sensing agent selection and sensing policy synthesis. However, since the policy search space is infinite, we adapt the classical Nemhauser-Wolsey argument to prove that the proposed IMAS$^2$ algorithm can provide a tight $(1 - 1/e)$-guarantee on the performance. Finally, we demonstrate the effectiveness of our approach in a multi-agent cooperative perception in a grid-world environment.

eess.SY↗

arXiv2025

Target Defense Using a Turret and Mobile Defender Team

A scenario is considered wherein a stationary, turn constrained agent (Turret) and a mobile agent (Defender) cooperate to protect the former from an adversarial mobile agent (Attacker). The Attacker wishes to reach the Turret prior to getting captured by either the Defender or Turret, if possible. Meanwhile, the Defender and Turret seek to capture the Attacker as far from the Turret as possible. This scenario is formulated as a differential game and solved using a geometric approach. Necessary and sufficient conditions for the Turret-Defender team winning and the Attacker winning are given. In the case of the Turret-Defender team winning equilibrium strategies for the min max terminal distance of the Attacker to the Turret are given. Three cases arise corresponding to solo capture by the Defender, solo capture by the Turret, and capture simultaneously by both Turret and Defender.

eess.SY↗

arXiv2025

Deceptive Path Planning: A Bayesian Game Approach

This paper investigates how an autonomous agent can transmit information through its motion in an adversarial setting. We consider scenarios where an agent must reach its goal while deceiving an intelligent observer about its destination. We model this interaction as a dynamic Bayesian game between a mobile Attacker with a privately known goal and a Defender who infers the Attacker's intent to allocate defensive resources effectively. We use Perfect Bayesian Nash Equilibrium (PBNE) as our solution concept and propose a computationally efficient approach to find it. In the resulting equilibrium, the Defender employs a simple Markovian strategy, while the Attacker strategically balances deception and goal efficiency by stochastically mixing shortest and non-shortest paths to manipulate the Defender's beliefs. Numerical experiments demonstrate the advantages of our PBNE-based strategies over existing methods based on one-sided optimization.

eess.SY↗

arXiv2024

Active Perception with Initial-State Uncertainty: A Policy Gradient Method

This paper studies the synthesis of an active perception policy that maximizes the information leakage of the initial state in a stochastic system modeled as a hidden Markov model (HMM). Specifically, the emission function of the HMM is controllable with a set of perception or sensor query actions. Given the goal is to infer the initial state from partial observations in the HMM, we use Shannon conditional entropy as the planning objective and develop a novel policy gradient method with convergence guarantees. By leveraging a variant of observable operators in HMMs, we prove several important properties of the gradient of the conditional entropy with respect to the policy parameters, which allow efficient computation of the policy gradient and stable and fast convergence. We demonstrate the effectiveness of our solution by applying it to an inference problem in a stochastic grid world environment.

eess.SY↗

arXiv2024

Heterogeneous Roles against Assignment Based Policies in Two vs Two Target Defense Game

In this paper, we consider a target defense game in which the attacker team seeks to reach a high-value target while the defender team seeks to prevent that by capturing them away from the target. To address the curse of dimensionality, a popular approach to solve such team-vs-team game is to decompose it into a set of one-vs-one games. Such an approximation assumes independence between teammates assigned to different one-vs-one games, ignoring the possibility of a richer set of cooperative behaviors, ultimately leading to suboptimality. In this paper, we provide teammate-aware strategies for the attacker team and show that they can outperform the assignment-based strategy, if the defenders still employ an assignment-based strategy. More specifically, the attacker strategy involves heterogeneous roles where one attacker actively intercepts a defender to help its teammate reach the target. We provide sufficient conditions under which such a strategy benefits the attackers, and we validate the results using numerical simulations.

eess.SY↗

arXiv2024

Deception in Differential Games: Information Limiting Strategy to Induce Dilemma

Can deception exist in differential games? We provide a case study for a Turret-Attacker differential game, where two Attackers seek to score points by reaching a target region while a Turret tries to minimize the score by aligning itself with the Attackers before they reach the target. In contrast to the original problem solved with complete information, we assume that the Turret only has partial information about the maximum speed of the Attackers. We investigate whether there is any incentive for the Attackers to move slower than their maximum speed in order to ``deceive'' the Turret into taking suboptimal actions. We first describe the existence of a dilemma that the Turret may face. Then we derive a set of initial conditions from which the Attackers can force the Turret into a situation where it must take a guess.

cs.GT↗

arXiv2024

Defending a Static Target Point with a Slow Defender

This paper studies a target-defense game played between a slow defender and a fast attacker. The attacker wins the game if it reaches the target while avoiding the defender's capture disk. The defender wins the game by preventing the attacker from reaching the target, which includes reaching the target and containing it in the capture disk. Depending on the initial condition, the attacker must circumnavigate the defender's capture disk, resulting in a constrained trajectory. This condition produces three phases of the game, which we analyze to solve for the game of kind. We provide the barrier surface that divides the state space into attacker-win and defender win regions, and present the corresponding strategies that guarantee win for each region. Numerical experiments demonstrate the theoretical results as well as the efficacy of the proposed strategies.

eess.SY↗

arXiv2024

Optimal Evasion from a Sensing-Limited Pursuer

This paper investigates a partial-information pursuit evasion game in which the Pursuer has a limited-range sensor to detect the Evader. Given a fixed final time, we derive the optimal evasion strategy for the Evader to maximize its distance from the pursuer at the end. Our analysis reveals that in certain parametric regimes, the optimal Evasion strategy involves a 'risky' maneuver, where the Evader's trajectory comes extremely close to the pursuer's sensing boundary before moving behind the Pursuer. Additionally, we explore a special case in which the Pursuer can choose the final time. In this scenario, we determine a (Nash) equilibrium pair for both the final time and the evasion strategy.

cs.GT↗

arXiv2023

The Eater and the Mover Game

This paper studies the idea of ``deception by motion'' through a two-player dynamic game played between a Mover who must retrieve resources at a goal location, and an Eater who can consume resources at two candidate goals. The Mover seeks to minimize the resource consumption at the true goal, and the Eater tries to maximize it. While the Mover has the knowledge about the true goal, the Eater cannot differentiate between the two candidates. Unlike existing works on deceptive motion control that measures the deceptiveness through the quality of inference made by a distant observer (an estimator), we incorporate their actions to directly measure the efficacy of deception through the outcome of the game. An equilibrium concept is then proposed without the notion of an estimator. We further identify a pair of equilibrium strategies and demonstrate that if the Eater optimizes for the worst-case scenario, hiding the intention (deception by ambiguity) is still effective, whereas trying to fake the true goal (deception by exaggeration) is not.

eess.SY↗

arXiv2023

Path Defense in Dynamic Defender-Attacker Blotto Games (dDAB) with Limited Information

We consider a path guarding problem in dynamic Defender-Attacker Blotto games (dDAB), where a team of robots must defend a path in a graph against adversarial agents. Multi-robot systems are particularly well suited to this application, as recent work has shown the effectiveness of these systems in related areas such as perimeter defense and surveillance. When designing a defender policy that guarantees the defense of a path, information about the adversary and the environment can be helpful and may reduce the number of resources required by the defender to achieve a sufficient level of security. In this work, we characterize the necessary and sufficient number of assets needed to guarantee the defense of a shortest path between two nodes in dDAB games when the defender can only detect assets within $k$-hops of a shortest path. By characterizing the relationship between sensing horizon and required resources, we show that increasing the sensing capability of the defender greatly reduces the number of defender assets needed to defend the path.

cs.GT↗

arXiv2023

Dynamic Adversarial Resource Allocation: the dDAB Game

This work introduces the dynamic Defender-Attacker Blotto (dDAB) game, extending the classical static Blotto game to a dynamic resource allocation setting over graphs. In the dDAB game, a defender is required to maintain numerical superiority against attacker resources across a set of key nodes in a connected graph. The engagement unfolds as a discrete-time game, where each player reallocates its resources in turn, with resources allowed to move at most one hop per time step. The primary goal is to determine the necessary and sufficient amount of defender resources required to guarantee sustained defense, along with the corresponding strategies. To address the central challenge arising from graph-constrained resource reallocation, we conduct a reachability analysis, starting with simplified settings where attacker resources act as a single cohesive group. We then extend the framework to allow attacker resources to split and merge arbitrarily, and construct defender strategies using superposition principles. A set-based dynamic programming algorithm is developed to compute the optimal strategies, as well as the minimum amount of defender resources to ensure successful defense. The effectiveness of our approach is demonstrated through numerical simulations and hardware experiments on the Georgia Tech Robotarium platform.

cs.MA↗

arXiv2022

Target Defense against Sequentially Arriving Intruders

We consider a variant of the target defense problem where a single defender is tasked to capture a sequence of incoming intruders. The intruders' objective is to breach the target boundary without being captured by the defender. As soon as the current intruder breaches the target or gets captured by the defender, the next intruder appears at a random location on a fixed circle surrounding the target. Therefore, the defender's final location at the end of the current game becomes its initial location for the next game. Thus, the players pick strategies that are advantageous for the current as well as for the future games. Depending on the information available to the players, each game is divided into two phases: partial information and full information phase. Under some assumptions on the sensing and speed capabilities, we analyze the agents' strategies in both phases. We derive equilibrium strategies for both the players to optimize the capture percentage using the notions of engagement surface and capture circle. We quantify the percentage of capture for both finite and infinite sequences of incoming intruders.

cs.RO↗

arXiv2022

A Constraint-Driven Approach to Line Flocking: The V Formation as an Energy-Saving Strategy

The study of robotic flocking has received significant attention in the past twenty years. In this article, we present a constraint-driven control algorithm that minimizes the energy consumption of individual agents and yields an emergent V formation. As the formation emerges from the decentralized interaction between agents, our approach is robust to the spontaneous addition or removal of agents to the system. First, we present an analytical model for the trailing upwash behind a fixed-wing UAV, and we derive the optimal air speed for trailing UAVs to maximize their travel endurance. Next, we prove that simply flying at the optimal airspeed will never lead to emergent flocking behavior, and we propose a new decentralized "anseroid" behavior that yields emergent V formations. We encode these behaviors in a constraint-driven control algorithm that minimizes the locomotive power of each UAV. Finally, we prove that UAVs initialized in an approximate V or echelon formation will converge under our proposed control law, and we demonstrate this emergence occurs in real-time in simulation and in physical experiments with a fleet of Crazyflie quadrotors.

cs.RO↗

arXiv2022

Guarding a Non-Maneuverable Translating Line with an Attached Defender

In this paper we consider a target-guarding differential game where the defender must protect a linearly translating line-segment by intercepting an attacker who tries to reach it. In contrast to common target-guarding problems, we assume that the defender is attached to the target and moves along with it. This assumption affects the defenders' maximum speed in inertial frame, which depends on the target's direction of motion. Zero-sum differential game of degree for both the attacker-win and defender-win scenarios are studied, where the payoff is defined to be the distance between the two agents at the time of game termination. We derive the equilibrium strategies and the Value function by leveraging the solution for the infinite-length target scenario. The zero-level set of this Value function provides the barrier surface that divides the state space into defender-win and attacker-win regions. We present simulation results to demonstrate the theoretical results.

eess.SY↗

arXiv2022

One Apollonius Circle is Enough for Many Pursuit-Evasion Games

This paper investigates obstacle-free simple motion pursuit-evasion problems where the pursuer is faster and game termination is point capture. It is well known that the interior of the Apollonius Circle (AC) is the evader's dominance region, however, it was unclear whether the evader could reach outside the initial AC without being captured. We construct a pursuit strategy that guarantees the capture of an evader within an arbitrarily close neighborhood of the initial AC. The pursuer strategy is derived by reformulating the game into a nonlinear control problem, and the guarantee holds against any admissible evader strategy. Our result implies that the evader can freely select the capture location, but only inside the initial AC. Therefore, a class of problems, including those where the payoff is determined solely based on the location of capture, are now trivial.

math.OC↗