Research Analyzer
← Back ICRA 2026

Inspection Planning under Execution Uncertainty

Shmuel David Alpert, Kiril Solovey, Itzik Klein, Oren Salzman

PDF

AI summary

Key figure (auto-extracted from paper)
IRIS-U2 enables reliable autonomous inspection under localization errors by integrating Monte Carlo sampling into path planning to provide statistical guarantees on coverage and safety.
inspection planning execution uncertainty Monte Carlo sampling UAV navigation path planning statistical guarantees

Problem

Existing inspection-planning algorithms ignore execution uncertainty, such as localization errors in urban environments, which causes missed inspection points and collisions. This gap hinders reliable autonomous inspection tasks like UAV bridge inspections.

Approach

The authors extend the deterministic IRIS algorithm by incorporating Monte Carlo sampling to estimate POI coverage probabilities and collision risks, using these statistical bounds to guide a refined search procedure that balances efficiency with safety guarantees.

Key results

  • IRIS-U2 algorithm for uncertainty-aware inspection planning
  • Statistical bounds on coverage, path length, and collision probability
  • Improved expected coverage and reduced collisions in UAV bridge inspections
  • Bounded suboptimal solutions for faster computation with preserved guarantees

Why it matters

Enables safer and more reliable autonomous UAV inspections in uncertain urban environments, critical for infrastructure maintenance and public safety.

Abstract

Autonomous inspection tasks require path-planning algorithms to efficiently gather observations from points of interest (POIs). However, localization errors in urban environments introduce execution uncertainty, posing challenges to successfully completing such tasks. Existing inspection-planning algorithms do not explicitly address this uncertainty, which can hinder their performance. To overcome this, we introduce IRIS-under uncertainty (IRIS-U2), an inspection-planning algorithm that provides statistical assurances regarding coverage, path length, and collision probability. Our approach builds upon IRIS —our framework for deterministic, highly efficient, and provably asymptotically-optimal framework. This extension adapts IRIS to uncertain settings using a refined search procedure that estimates POI coverage probabilities through Monte Carlo (MC) sampling. We demonstrate IRIS-U2 through a case study on bridge inspections, achieving improved expected coverage, reduced collision probability, and increasingly precise statistical guarantees as MC samples grow. Additionally, we explore bounded suboptimal solutions to reduce computation time while preserving statistical assurances.

Index terms

Inspection planning under uncertainty Motion and Path Planning Aerial Systems: Applications Collision Avoidance

Related papers