Monitoring autonomous persistent surveillance missions using invariance

2026-05-07

AI Summary

This work addresses the challenge of verifying correctness and safety in real time for autonomous robots performing persistent surveillance tasks, where black-box autonomy complicates online validation. The authors propose a compositional runtime monitoring approach that partitions the environment into a finite set of regions, modeled as a state-dependent hybrid system. Regional uncertainties are captured using linear parameter-varying dynamics, enabling the offline computation of low-dimensional invariant sets for each region. At runtime, the method verifies the joint satisfaction of these invariants. Under an assumption of regional independence, the approach provides sound and complete guarantees for global invariance properties. Experimental validation on a physical robot navigating a maze demonstrates the method’s effectiveness and practical applicability.

Abstract

This paper studies runtime monitoring for persistent surveillance by autonomous robots when the autonomy stack is a black box. The environment is partitioned into finitely many parts, each carrying an uncertainty state that decreases when observed and increases otherwise. We model the closed loop as a state-dependent hybrid system with linear parameter varying dynamics and design a monitor based on an invariant computed offline. As this invariant is typically hard to obtain for large to-be-surveyed spaces, we propose a compositional monitor obtained by decentralized computation of low-dimensional invariant sets for each uncertainty region, and checking their conjunction online. Under common independence assumptions, the compositional monitor is sound and complete with respect to the full-system invariant. The approach is applied in a case study with a real robot persistently monitoring a labyrinth, emphasizing its applicability in practice.

Problem

• runtime monitoring
• persistent surveillance
• autonomous robots
• black-box autonomy
• invariance

Innovation

• runtime monitoring
• invariant sets
• compositional verification
• hybrid systems
• persistent surveillance