analiza aiqbench
This research introduces ARMOR, a model-free reinforcement learning controller designed to maintain UAV safety against adversarial sensor attacks like GPS spoofing by learning robust latent state representations. The method employs a two-stage training framework where an initial teacher encoder generates attack-aware states that are later approximated by a student encoder for practical deployment without privileged data. By improving generalization to unseen threats and reducing the need for iterative adversarial training, this work advances the security of autonomous robotic systems in hostile environments.
ARMOR: Robust Reinforcement Learning-based Control for UAVs under Physical Attacks
arXiv:2506.22423v2 Announce Type: replace-cross Abstract: Unmanned Aerial Vehicles (UAVs) depend on onboard sensors for perception, navigation, and control. However, these sensors are susceptible to physical attacks, such as GPS spoofing, that can corrupt state estimates and lead to unsafe behavior. While reinforcement learning (RL) offers adaptive control capabilities, existing safe RL methods are ineffective against such attacks. We present ARMOR (Adaptive Robust Manipulation-Optimized State Representations), an attack-resilient, model-free RL controller that enables robust UAV operation under adversarial sensor manipulation. Instead of relying on raw sensor observations, ARMOR learns a robust latent representation of the UAV's physical state via a two-stage training framework. In the first stage, a teacher encoder, trained with privileged attack information, generates attack-aware latent states for RL policy training. In the second stage, a student encoder is trained via supervised learning to approximate the teacher's latent states using only historical sensor data, enabling real-world deployment without privileged information. Our experiments show that ARMOR outperforms conventional methods, ensuring UAV safety. Additionally, ARMOR improves generalization to unseen attacks and reduces training cost by eliminating the need for iterative adversarial training.
Abstract:Unmanned Aerial Vehicles (UAVs) depend on onboard sensors for perception, navigation, and control. However, these sensors are susceptible to physical attacks, such as GPS spoofing, that can corrupt state estimates and lead to unsafe behavior. While reinforcement learning (RL) offers adaptive control capabilities, existing safe RL methods are ineffective against such attacks. We present ARMOR (Adaptive Robust Manipulation-Optimized State Representations), an attack-resilient, model-free RL controller that enables robust UAV operation under adversarial sensor manipulation. Instead of relying on raw sensor observations, ARMOR learns a robust latent representation of the UAV's physical state via a two-stage training framework. In the first stage, a teacher encoder, trained with privileged attack information, generates attack-aware latent states for RL policy training. In the second stage, a student encoder is trained via supervised learning to approximate the teacher's latent states using only historical sensor data, enabling real-world deployment without privileged information. Our experiments show that ARMOR outperforms conventional methods, ensuring UAV safety. Additionally, ARMOR improves generalization to unseen attacks and reduces training cost by eliminating the need for iterative adversarial training.
Machine Learning (cs.LG); Cryptography and Security (cs.CR); Robotics (cs.RO)
arXiv:2506.22423 [cs.LG]
(or arXiv:2506.22423v2 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2506.22423 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 274 cuvinte
Subiecte conexe
Curriculum-Based Adversarial Heterogeneous Agent Reinforcement Learning for Autonomous Quad-Copter Landing in Maritime Settings
arXiv · 14 sept. 2026
Intelligent Multi-UAV Navigation in ITNTNs: A Hierarchical LLM Approach
arXiv · 23 iul. 2026
Latent reasoning architectures would likely undermine CoT, our strongest oversight tool
AI Alignment Forum · 23 sept. 2026
Continual learning might make your blocking monitors nearly useless
AI Alignment Forum · 24 sept. 2026