Shield Synthesis for LTL Modulo Theories

2024-06-06

arXiv.org

AI Summary

Existing shielding techniques are restricted to propositional temporal logics (e.g., LTL) and thus unable to express complex safety specifications involving data constraints and continuous dynamics prevalent in safety-critical systems. Method: This paper pioneers the extension of shielding to LTL modulo Theories (LTL^MT), enabling modeling of hybrid discrete-continuous behaviors and parameterized predicates. We propose an automated shield synthesis framework integrating reactive synthesis modulo theories, SMT solving, and LTL controller synthesis to generate verifiable and executable runtime monitors. Contribution/Results: Our approach guarantees formal correctness and real-time enforceability. Experimental evaluation on robotic control and cyber-physical systems demonstrates high-precision detection and mitigation of violations against joint temporal–data specifications, achieving both low latency and strong enforcement fidelity.

Technology Category

• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories
• Natural Language Processing: Safety and Robustness
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Security and Privacy: Large-scale security measurements
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

In recent years, Machine Learning (ML) models have achieved remarkable success in various domains. However, these models also tend to demonstrate unsafe behaviors, precluding their deployment in safety-critical systems. To cope with this issue, ample research focuses on developing methods that guarantee the safe behaviour of a given ML model. A prominent example is shielding which incorporates an external component (a ``shield'') that blocks unwanted behavior. Despite significant progress, shielding suffers from a main setback: it is currently geared towards properties encoded solely in propositional logics (e.g., LTL) and is unsuitable for richer logics. This, in turn, limits the widespread applicability of shielding in many real-world systems. In this work, we address this gap, and extend shielding to LTL modulo theories, by building upon recent advances in reactive synthesis modulo theories. This allowed us to develop a novel approach for generating shields conforming to complex safety specifications in these more expressive, logics. We evaluated our shields and demonstrate their ability to handle rich data with temporal dynamics. To the best of our knowledge, this is the first approach for synthesizing shields for such expressivity.

Problem

• Extend shielding to LTL modulo theories
• Generate shields for complex safety specifications
• Handle rich data with temporal dynamics

Innovation

• Extends shielding to LTL modulo theories
• Uses reactive synthesis modulo theories
• Generates shields for complex safety specifications