A Probabilistic Eulerian Traffic Model for the Coordination of Multiple AGVs in Automatic Warehouses

The co-ordination of multiple automated guided vehicles (AGVs) is one of the main issues to address for the implementation of an efficient autonomous warehouse. Traffic jams are highly undesirable and, therefore, the motion of the AGVs should be planned by considering the current and future state of the fleet. This letter proposes a probabilistic model of the traffic in an autonomous warehouse to predict the evolution of possibly congested areas. Such a model is then exploited for building a predictive planner that is embedded in the traffic manager recently proposed in [1] to explicitely consider the evolution of the traffic up to a given horizon and to increase the efficiency of the fleet of AGVs in terms of delivery time. The proposed traffic manager is validated by means of comparative simulations on real plants.