No Need to Stop the Fleet: Localized Anomaly Isolation via Dead Zones in AGV Fleets

Natalia Ogorelysheva; Falk Howar; Jian-Jia Chen

cs.ROOct 6, 2026

No Need to Stop the Fleet: Localized Anomaly Isolation via Dead Zones in AGV Fleets

Authors: Natalia Ogorelysheva, Falk Howar, Jian-Jia Chen

Organizations: TU Dortmund University Dortmund, Germany · Fraunhofer Institute for Software and Systems Engineering ISST Dortmund, Germany · RWTH Aachen University Aachen, Germany

Abstract

Automated Guided Vehicle (AGV) fleets in production and logistics environments rely on a system-wide emergency stop to handle local anomalies such as malfunctions, hazards, or contaminated areas. While safe, this approach halts the entire fleet even when only a small area is affected, causing unnecessary downtime. This paper proposes dead zones --- a formally defined localized isolation approach that serves as an additional fail-operational mechanism for situations where a full system-wide halt is not strictly needed. Rather than shutting down the entire fleet, a dead zone isolates the anomalous area, allowing unaffected robots to continue operating normally. We introduce two operational strategies: basic dead zone handling, in which affected robots enter a dead zone and stop, and dead zone handling with escaping, in which robots that are sufficiently far from a dead zone reroute around it. For escaping on counterclockwise trajectories we prove collision-freedom with a bounded arrival delay for all escaping robots, provided a minimum pairwise separation is kept between robots. Experimental evaluation confirms the theoretical guarantees and characterizes the throughput tradeoffs between all three approaches: full emergency stop, basic dead zone handling, and dead zone handling with escaping.

Figures & tables

Panel of Figure 1 — Figure 1 : Five robots operating on a grid before and after a dead zone is announced.

Panel of Figure 2 — Figure 2 : Simulation results: grid $50{\times}50$ , $T=1000$ , $30$ seeds per configuration. Fleet size sweep (DZ $1{\times}1$ , fixed). All error bars show mean $\pm$ std.

Panel of Figure 3 — Figure 3 : Simulation results: grid $50{\times}50$ , $T=1000$ , $30$ seeds per configuration. DZ size sweep ( $N=15$ , fixed). All error bars show mean $\pm$ std.

Explore similar work

HiRAD: A Flexible Large-Scale AGV Routing System
Sep 9, 2026Yunjie Huang, Ruizhong Wu, Mengxuan Zhang +3Vehicle Routing Problem Hierarchical Reinforcement Learning
A traffic management system for large and heterogeneous vehicles in narrow industrial environments
Sep 9, 2026Alessandro Bonetti, Silvia Proia, Simone Guidetti +1Intelligent Transportation Multi-Agent Path Finding
Department of Sciences and Methods for Engineering (DISMI), University of Modena and Reggio Emilia, Italy · Gruppo TecnoFerrari S.p.a. con socio unico, Italy
PREVENT-JACK: Context Steering for Swarms of Long Heavy Articulated Vehicles
Apr 23, 2026Adrian Baruck, Michael Dubé, Christoph Steup +1Swarms Collision Avoidance
1*Chair of Computational Intelligence, Otto-von-Guericke-University, Universitätsplatz 2, Magdeburg, 39106, Germany. · 2Fraunhofer Institute for Transportation and Infrastructure Systems2026 IVI, Zeunerstr. 38, Dresden, 01069, Germany.

Sep 9, 2026, cs.ROJ/K move · Enter open · S save

HiRAD: A Flexible Large-Scale AGV Routing System

Yunjie Huang, Ruizhong Wu, Mengxuan Zhang, Frodo Kin Sun Chan+2

Vehicle Routing Problem Hierarchical Reinforcement Learning Offline Reinforcement Learning Model-Based Multi-Agent Systems Spatiotemporal Representations

Open paper AskPDF arXiv

Automatic Guided Vehicles (AGVs) substantially boost warehouse throughput, but routing large-scale AGV fleets remains challenging. Classical Multi-Agent Pathfinding solvers suffer from exploding combinatorial complexity and super-quadratic runtime, while relying on idealized grid or piecewise-linear motion models that mismatch real-world kinematics. Recent Reinforcement Learning (RL) solutions improve flexibility via decentralized agent policies but depend on discretized spatiotemporal representations, require millions of episodes to converge, and incur full-map observation at every step, which leads to large models, slow convergence, and high inference latency that violates real-time industrial control constraints. To address these bottlenecks, we propose HiRAD, a hierarchical RL framework for continuous-space AGV routing with real-time guarantees: (1) a step-level spatiotemporal representation that translates continuous motion into a differentiable RL problem, (2) a hierarchical strategy that splits heading choice from velocity control to reduce the action space, and (3) an asynchronous event-driven decision pipeline that lowers inference complexity from O(n^2) to O(n) and cuts per-step latency by as much as 71 percent. Across random graphs and two warehouse maps, HiRAD reduces makespan by 45 percent to 63 percent and shortens end-to-end runtime.

Abstract

Explore similar work

HiRAD: A Flexible Large-Scale AGV Routing System

A traffic management system for large and heterogeneous vehicles in narrow industrial environments

PREVENT-JACK: Context Steering for Swarms of Long Heavy Articulated Vehicles