Harnessing Embodied Agents: Runtime Governance for Policy-Constrained Execution

2026-04-09

AI Summary

This work addresses the critical need for effective governance and control of embodied intelligent agents even after they have been granted execution privileges. It establishes runtime governance as a foundational system-level challenge and introduces a novel framework that decouples cognitive reasoning from execution oversight. By externalizing a dedicated governance layer, the approach enables policy verification, capability access control, real-time execution monitoring, automated rollback, and human-in-the-loop intervention. The architecture explicitly delineates control boundaries among the agent, capability modules, and the governance layer, positioning policy constraints as a core design principle for embodied systems. In 1,000 randomized simulations, the method successfully intercepted 96.2% of unauthorized operations, reduced unsafe continuations under runtime drift from 100% to 22.2%, and achieved a compliance recovery rate of 91.4%, significantly outperforming all baselines (p < 0.001).

Abstract

Embodied agents are evolving from passive reasoning systems into active executors that interact with tools, robots, and physical environments. Once granted execution authority, the central challenge becomes how to keep actions governable at runtime. Existing approaches embed safety and recovery logic inside the agent loop, making execution control difficult to standardize, audit, and adapt. This paper argues that embodied intelligence requires not only stronger agents, but stronger runtime governance. We propose a framework for policy-constrained execution that separates agent cognition from execution oversight. Governance is externalized into a dedicated runtime layer performing policy checking, capability admission, execution monitoring, rollback handling, and human override. We formalize the control boundary among the embodied agent, Embodied Capability Modules (ECMs), and runtime governance layer, and validate through 1000 randomized simulation trials across three governance dimensions. Results show 96.2% interception of unauthorized actions, reduction of unsafe continuation from 100% to 22.2% under runtime drift, and 91.4% recovery success with full policy compliance, substantially outperforming all baselines (p<0.001). By reframing runtime governance as a first-class systems problem, this paper positions policy-constrained execution as a key design principle for embodied agent systems.

Problem

• embodied agents
• runtime governance
• policy-constrained execution
• execution oversight
• safety compliance

Innovation

• runtime governance
• policy-constrained execution
• embodied agents
• execution oversight
• capability admission