Managed Autonomy at Runtime: Gear-Based Safety and Governance for Single- and Multi-Agent Cyber-Physical Systems

2026-06-30

AI Summary

Autonomous agents operating without continuous human oversight are prone to safety violations and behavioral instability. This work proposes a discrete-time control system that, for the first time, integrates a five-tier execution mechanism—comprising Gobs, Gsug, Gplan, Gexec, and Gint—with the SMART governance lifecycle to decouple action control from autonomous governance while providing formal safety guarantees. The framework incorporates utility-gated scheduling, event-triggered fallback, consensus gating, and collective Lyapunov analysis to ensure runtime safety in both single- and multi-agent cyber-physical systems. Experimental validation on a UR5 triple-arm robotic platform demonstrates a 99.6% anomaly detection rate (versus 2.1% for the baseline), a 3.5× reduction in detection latency, and the generation of verifiable safety certificates over the physical workspace.

Abstract

Autonomous agents, whether LLM-driven software agents or robotic physical agents, face a common class of failure modes when operating without continuous human oversight: safety violations from unverified actions, behavioral instability from unconstrained loops, and continuity loss from unhandled error states. We develop \system{}, a discrete-time control system that combines five execution gears (\Gobs{}, \Gsug{}, \Gplan{}, \Gexec{}, \Gint{}) with utility-gated dispatch and event-driven fallback. For the single-agent case, we prove monotonic stability, execution safety, eventual stabilization, fallback completeness, and equivalence to a gear-constrained Markov decision process. For multi-agent cyber-physical systems (CPS), we apply the established \smart{} managed-autonomy lifecycle and map runtime evidence into its four governance states (\Stable{}/\Meta{}/\Assisted{}/\Regulated{}). Consensus gating, swarm-level Lyapunov analysis, per-agent gear authority, and rendezvous control provide distributed safety and stability guarantees, including zero collision under the stated assumptions. We evaluate the resulting runtime on a three-agent UR5 robotic assembly cell using fault magnitudes calibrated from the NIST \emph{Degradation Measurement of Robot Arm Position Accuracy} dataset across 10,000 Monte Carlo episodes. It achieves a 99.6\% anomaly detection rate versus 2.1\% for the single-agent baseline, reduces detection latency by $3.5\times$, and supplies a formal physical-workspace safety certificate. The execution gears act as micro-level permissions beneath the \smart{} runtime governance states, separating action control from autonomy governance.

Problem

• Managed Autonomy
• Safety Violations
• Behavioral Instability
• Continuity Loss
• Cyber-Physical Systems

Innovation

• Managed Autonomy
• Execution Gears
• Runtime Governance
• Multi-Agent CPS
• Formal Safety Certification