Learning Contextual Runtime Monitors for Safe AI-Based Autonomy

2026-01-28

AI Summary

This work addresses the significant performance degradation of AI controllers in unknown environments and the challenge of simultaneously ensuring safety and contextual adaptability with conventional ensemble methods. The authors formulate safe AI control integration as a contextual monitoring problem and propose a context-aware runtime monitoring framework that dynamically selects the most suitable controller for the current environment—rather than fusing outputs—to theoretically guarantee system safety while effectively leveraging controller diversity. The monitor employs a contextual multi-armed bandit algorithm to learn and perform context-sensitive controller scheduling. Evaluated in two autonomous driving simulation scenarios, the approach substantially improves both system safety and task performance compared to non-contextual baselines.

Technology Category

• Philosophy and Ethics of AI: Safety, Robustness & Trustworthiness
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Multiagent Systems: Adversarial Agents

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Fairness and ethical considerations in crowd work and in human-in-the-loop AI systems
• Security and Privacy: Security and privacy of machine learning and AI applications

Abstract

We introduce a novel framework for learning context-aware runtime monitors for AI-based control ensembles. Machine-learning (ML) controllers are increasingly deployed in (autonomous) cyber-physical systems because of their ability to solve complex decision-making tasks. However, their accuracy can degrade sharply in unfamiliar environments, creating significant safety concerns. Traditional ensemble methods aim to improve robustness by averaging or voting across multiple controllers, yet this often dilutes the specialized strengths that individual controllers exhibit in different operating contexts. We argue that, rather than blending controller outputs, a monitoring framework should identify and exploit these contextual strengths. In this paper, we reformulate the design of safe AI-based control ensembles as a contextual monitoring problem. A monitor continuously observes the system's context and selects the controller best suited to the current conditions. To achieve this, we cast monitor learning as a contextual learning task and draw on techniques from contextual multi-armed bandits. Our approach comes with two key benefits: (1) theoretical safety guarantees during controller selection, and (2) improved utilization of controller diversity. We validate our framework in two simulated autonomous driving scenarios, demonstrating significant improvements in both safety and performance compared to non-contextual baselines.

Problem

• AI-based autonomy
• runtime monitoring
• contextual safety
• controller ensemble
• cyber-physical systems

Innovation

• contextual runtime monitor
• AI-based autonomy
• contextual multi-armed bandits
• safe control ensembles
• controller selection