# Time-Constrained Intelligent Adversaries for Automation Vulnerability Testing: A Multi-Robot Patrol Case Study

arXiv: https://arxiv.org/abs/2509.11971 · HTML: https://mlsanity.com/paper/2509.11971/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 40 references that are in the map.
- First posted on arXiv 15 Sep 2025.

## Builds on (most cited first, 12 of 12)

- [Graph Neural Networks for Decentralized Multi-Robot Path Planning](https://mlsanity.com/paper/1912.06095/) (2019)
- [Learning Decentralized Controllers for Robot Swarms with Graph Neural Networks](https://mlsanity.com/paper/1903.10527/) (2019)
- [Multi-Robot Collaborative Perception with Graph Neural Networks](https://mlsanity.com/paper/2201.01760/) (2022)
- [Embodied Red Teaming for Auditing Robotic Foundation Models](https://mlsanity.com/paper/2411.18676/) (2024)
- [Multi-Robot Coverage and Exploration using Spatial Graph Neural Networks](https://mlsanity.com/paper/2011.01119/) (2020)
- [Predictive Red Teaming: Breaking Policies Without Breaking Robots](https://mlsanity.com/paper/2502.06575/) (2025)
- [Multi-Robot Adversarial Patrolling: Facing a Full-Knowledge Opponent](https://mlsanity.com/paper/1401.3903/) (2014)
- [Approximation Algorithms for Multi-Robot Patrol-Scheduling with Min-Max Latency](https://mlsanity.com/paper/2005.02530/) (2020)
- [On Cyclic Solutions to the Min-Max Latency Multi-Robot Patrolling Problem](https://mlsanity.com/paper/2203.07280/) (2022)
- [Markov Chain-Based Stochastic Strategies for Robotic Surveillance](https://mlsanity.com/paper/2008.09050/) (2020)
- [Towards a Robotic Intrusion Prevention System: Combining Security and Safety in Cognitive Social Robots](https://mlsanity.com/paper/2407.06669/) (2024)
- [Lightweight Decentralized Neural Network-Based Strategies for Multi-Robot Patrolling](https://mlsanity.com/paper/2412.11916/) (2024)

## Abstract (arXiv)

Simulating hostile attacks of physical autonomous systems can be a useful tool to examine their robustness to attack and inform vulnerability-aware design. In this work, we examine this through the lens of multi-robot patrol, by presenting a machine learning-based adversary model that observes robot patrol behavior in order to attempt to gain undetected access to a secure environment within a limited time duration. Such a model allows for evaluation of a patrol system against a realistic potential adversary, offering insight into future patrol strategy design. We show that our new model outperforms existing baselines, thus providing a more stringent test, and examine its performance against multiple leading decentralized multi-robot patrol strategies.