Harnessing Embodied Agents: Runtime Governance for Policy-Constrained Execution
2026-04-09 • Robotics
Robotics
AI summaryⓘ
The authors explain that robots and smart agents that act in the real world need better ways to control what they do during their actions. Instead of mixing safety rules inside the agent, they suggest having a separate system that watches and checks everything the agent tries to do. They tested this setup and found it stops most bad actions and helps fix problems safely much better than older methods. This approach treats safety checks as an important part of the whole system, not just something extra.
Embodied agentsRuntime governancePolicy-constrained executionExecution monitoringSafety enforcementCapability modulesRollback handlingHuman overrideSimulation trials
Authors
Xue Qin, Simin Luan, John See, Cong Yang, Zhijun Li
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.