Hierarchical Traffic Management of Multi-AGV Systems With Deadlock Prevention Applied to Industrial Environments

This paper concerns the coordination and traffic management of a group Automated Guided Vehicles (AGVs) moving in real industrial scenario, such as an automated factory or warehouse. The proposed methodology is based on three-layer control architecture, which described follows: 1) Top Layer (or Topological Layer) allows to model vehicles among different areas environment; 2) Middle path planner compute sensitive for each vehicle; 3) Bottom Roadmap defines final routes be followed by vehicle coordinates AGVs over time. In we describe strategy propose, executed once are computed has aim prevent congestions, collisions deadlocks. algorithm exploits novel deadlock prevention approach time-expanded graphs. Moreover, presented architecture aims at grounding theoretical methods application facing typical practical issues graphs difficulties (load/unload locations, weak connections,), predefined roadmap (constrained plant layout), errors, dynamical obstacles, etc. this propose flexible robust multi-AGVs traffic-aware management. algorithm, does not rely ad hoc assumptions rules, deadlocks deal with delays motion errors. Note Practitioners —This connections, ),