Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Position: Certifiable State Integrity in Cyber-Physical Systems -- Why Modular Sovereignty Solves the Plasticity-Stability Paradox

Enzo Nicolás Spotorno, Antônio Augusto Medeiros Fröhlich

TL;DR

The paper addresses the certifiability gap of monolithic foundation models in safety-critical CPS by defining State Integrity and proposing HYDRA, a Modular Sovereignty architecture that uses a library of frozen regime-specific specialists blended by an uncertainty-aware Governor. Outputs are constrained to convex combinations of valid local operators, enabling regime-conditional validity and modular certification via LPV-like analysis and geometric safety envelopes such as RPI Zonotopes. Offline Constitution grows the specialist library with physics- and data-derived experts, while online governance arbiters mixing weights and triggers safe handovers through an Integrity Triangle (Governance Health, Uncertainty, Regime Stability). The approach aims to deliver low-latency, auditable adaptation across lifecycle CPS while supporting formal verification and compliance with safety standards, demonstrated conceptually through vehicle dynamics and related CPS domains.

Abstract

The machine learning community has achieved remarkable success with universal foundation models for time-series and physical dynamics, largely overcoming earlier approximation barriers in smooth or slowly varying regimes through scale and specialized architectures. However, deploying these monolithic models in safety-critical Cyber-Physical Systems (CPS), governed by non-stationary lifecycle dynamics and strict reliability requirements, reveals persistent challenges. Recent evidence shows that fine-tuning time-series foundation models induces catastrophic forgetting, degrading performance on prior regimes. Standard models continue to exhibit residual spectral bias, smoothing high-frequency discontinuities characteristic of incipient faults, while their opacity hinders formal verification and traceability demanded by safety standards (e.g., ISO 26262, IEC 61508). This position paper argues that the plasticity-stability paradox cannot be fully resolved by global parameter updates (whether via offline fine-tuning or online adaptation). Instead, we advocate a Modular Sovereignty paradigm: a library of compact, frozen regime-specific specialists combined via uncertainty-aware blending, which we term "HYDRA" (Hierarchical uncertaintY-aware Dynamics for Rapidly-Adapting systems). This paradigm ensures regime-conditional validity, rigorous disentanglement of aleatoric and epistemic uncertainties, and modular auditability, offering a certifiable path for robust state integrity across the CPS lifecycle.

Position: Certifiable State Integrity in Cyber-Physical Systems -- Why Modular Sovereignty Solves the Plasticity-Stability Paradox

TL;DR

The paper addresses the certifiability gap of monolithic foundation models in safety-critical CPS by defining State Integrity and proposing HYDRA, a Modular Sovereignty architecture that uses a library of frozen regime-specific specialists blended by an uncertainty-aware Governor. Outputs are constrained to convex combinations of valid local operators, enabling regime-conditional validity and modular certification via LPV-like analysis and geometric safety envelopes such as RPI Zonotopes. Offline Constitution grows the specialist library with physics- and data-derived experts, while online governance arbiters mixing weights and triggers safe handovers through an Integrity Triangle (Governance Health, Uncertainty, Regime Stability). The approach aims to deliver low-latency, auditable adaptation across lifecycle CPS while supporting formal verification and compliance with safety standards, demonstrated conceptually through vehicle dynamics and related CPS domains.

Abstract

The machine learning community has achieved remarkable success with universal foundation models for time-series and physical dynamics, largely overcoming earlier approximation barriers in smooth or slowly varying regimes through scale and specialized architectures. However, deploying these monolithic models in safety-critical Cyber-Physical Systems (CPS), governed by non-stationary lifecycle dynamics and strict reliability requirements, reveals persistent challenges. Recent evidence shows that fine-tuning time-series foundation models induces catastrophic forgetting, degrading performance on prior regimes. Standard models continue to exhibit residual spectral bias, smoothing high-frequency discontinuities characteristic of incipient faults, while their opacity hinders formal verification and traceability demanded by safety standards (e.g., ISO 26262, IEC 61508). This position paper argues that the plasticity-stability paradox cannot be fully resolved by global parameter updates (whether via offline fine-tuning or online adaptation). Instead, we advocate a Modular Sovereignty paradigm: a library of compact, frozen regime-specific specialists combined via uncertainty-aware blending, which we term "HYDRA" (Hierarchical uncertaintY-aware Dynamics for Rapidly-Adapting systems). This paradigm ensures regime-conditional validity, rigorous disentanglement of aleatoric and epistemic uncertainties, and modular auditability, offering a certifiable path for robust state integrity across the CPS lifecycle.
Paper Structure (5 sections, 2 figures, 1 table)

This paper contains 5 sections, 2 figures, 1 table.

Table of Contents

  1. Introduction: From Approximation Breakthroughs to the Lifecycle Certifiability Barrier
  2. The Theoretical Core: Persisting Barriers to Certifiable Lifecycle Modeling
  3. HYDRA: An Architectural Pattern for Certifiable Lifecycle Adaptation
  4. Alternative Perspectives and Rebuttals
  5. Discussion and Future Directions: Toward Certifiable Physical Learning

Figures (2)

  • Figure 1: Simplex Integration. The Governor isolates the AI (QM) from the Safety Core (ASIL D). Specialists provide estimation; the Governor and fallback form the high-assurance safety channel. Figure crafted with the help of a GenAI Image Model.
  • Figure 2: The Architecture of Sovereignty. Illustrated via the Vehicle Dynamics case study. The framework separates the offline Constitution (Library of Specialists) from the online Governance (in this example, implementing a Bayesian method for uncertainty-quantification, such as variational inference), allowing the system to switch "jurisdictions" rather than retraining weights during regime drifts.