Threat-Oriented Digital Twinning for Security Evaluation of Autonomous Platforms

2026-04-28

AI Summary

Current open research struggles to effectively evaluate the cybersecurity of learning-driven systems under realistic autonomous platforms, adversarial environments, and degraded communication conditions. This work proposes a threat-informed digital twin approach that constructs an open-source, modular autonomous agent with decoupled perception, decision-making, and monitoring components, enabling observable and controllable security testing. It innovatively translates threat modeling into reproducible design patterns to systematically support stress testing against deception, replay, malformed inputs, perceptual degradation, and adversarial machine learning attacks. By integrating multimodal confidence-gated perception, explicit trust boundaries, and runtime safety mechanisms, the framework accommodates typical constraints of aerospace systems—such as limited computation, intermittent links, and perceptual uncertainty—providing a general, implementable platform for secure autonomous systems research across ground, aerial, and space domains.

Abstract

Open, unclassified research on secure autonomy is constrained by limited access to operational platforms, contested communications infrastructure, and representative adversarial test conditions. This paper presents a threat-oriented digital twinning methodology for cybersecurity evaluation of learning-enabled autonomous platforms. The approach is instantiated as an open-source, modular twin of a representative autonomy stack with separated sensing, autonomy, and supervisory-control functions; confidence-gated multi-modal perception; explicit command and telemetry trust boundaries; and runtime hold-safe behavior. The contribution is methodological: a reproducible design pattern that translates threat analysis into observable, controllable tests for spoofing, replay, malformed-input injection, degraded sensing, and adversarial ML stress. Although the implemented proxy is ground based, the architecture is intentionally framed around stack elements shared with UAV and space systems, including constrained onboard compute, intermittent or high-latency links, probabilistic perception, and mission-critical recovery behavior. The result is an implementable research scaffold for dependable and secure autonomy studies across UAV and space domains.

Problem

• Digital Twin
• Autonomous Platforms
• Cybersecurity Evaluation
• Adversarial Testing
• Threat Analysis

Innovation

• threat-oriented digital twin
• secure autonomy
• adversarial machine learning
• trust boundaries
• modular autonomy stack