The regulation of the AGVs flow along the lanes inside the warehouse is analyzed as the urban traffic flow using a SUMO urban mobility simulator. This tool is used to generate data related to the AGV movement, and a reinforcement learning scheme, combining agent-based modeling and VLC queuing/request/response behaviors, effectively schedules routes. This provides efficient travel and avoids crowded regions. The demonstration of this proof-of-concept is supported on the evaluation of travel time and traffic flows. |
|