Robust and Efficient AI-Based Attack Recovery in Autonomous Drones

2025-05-20

AI Summary

Autonomous UAVs lack lightweight, real-time, offline recovery capabilities following cyberattacks. Method: This paper proposes an edge-oriented autonomous attack recovery architecture that—novelty—integrates commonsense reasoning into the UAV security recovery loop, combining a lightweight knowledge graph with rule-augmented neuro-symbolic reasoning. The system is deployed on the Jetson AGX Orin platform via model pruning, quantization, and protection within a Trusted Execution Environment (TEE). Contribution/Results: It operates entirely offline—without cloud dependency—and enables semantic-level autonomous decision-making. Recovery decision latency is under 80 ms, and the architecture achieves >92% recovery success rate against six representative attacks—including GPS spoofing and communication hijacking—demonstrating strong robustness, real-time performance, and security assurance.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Natural Language Processing: Safety and Robustness
• Machine Learning: Hardware-aware ML

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

We introduce an autonomous attack recovery architecture to add common sense reasoning to plan a recovery action after an attack is detected. We outline use-cases of our architecture using drones, and then discuss how to implement this architecture efficiently and securely in edge devices.

Problem

• Autonomous attack recovery in drones
• Common sense reasoning for recovery actions
• Efficient secure implementation on edge devices

Innovation

• Autonomous attack recovery architecture for drones
• Common sense reasoning for recovery actions
• Efficient secure implementation on edge devices