Deceptive Path Planning: A Bayesian Game Approach

2025-06-16

AI Summary

This paper addresses deceptive path planning for autonomous agents operating in adversarial environments: an agent must reach its true goal while simultaneously misleading observers’ inferences about its intent through its motion trajectory. We formulate this as a dynamic Bayesian game between an attacker (with a private goal) and a defender (an intent inferer), and introduce, for the first time, the Perfect Bayesian Nash Equilibrium (PBNE) to compute their jointly optimal strategies. Our method integrates a Markovian defensive policy with a stochastic mixed-path mechanism, enabling the attacker to actively perturb the observer’s belief state. Numerical experiments demonstrate that the PBNE strategy significantly improves goal-reaching success rates and reduces defender resource misallocation compared to unilateral optimization approaches. This work establishes a provably optimal game-theoretic framework for intent-hiding and adversarial navigation.

Technology Category

• Multiagent Systems: Adversarial Agents
• Search and Optimization: Adversarial Search
• Game Theory and Economic Paradigms: Adversarial Learning

Application Category

• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

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.

Problem

• Autonomous agent motion in adversarial settings
• Deceiving observer about agent's true destination
• Dynamic Bayesian game for deceptive path planning

Innovation

• Dynamic Bayesian game models adversarial motion planning
• Perfect Bayesian Nash Equilibrium ensures efficient computation
• Stochastic path mixing balances deception and goal efficiency