Development of an intelligent agent-based AGV controller for a flexible manufacturing system

Abstract

Automated guided vehicles (AGVs) are the most flexible means to transport materials among workstations of a flexible manufacturing system. Complex issues associated with the design of AGV control of these systems are conflict-free shortest path, minimum time motion planning and deadlock avoidance. This research presents an intelligent agent-based framework to overcome the inefficacies associated with the aforementioned issues. Proposed approach describes the operational control of AGVs by integrating different activities such as path generation, journey time enumeration, collision and deadlock identification, waiting node location and its time estimation, and decision making on the selection of the conflict-free shortest feasible path. It represents efficient algorithms and rules associated with each agent for finding the conflict-free minimum time motion planning of AGVs, which are needed to navigate unidirectional and bidirectional flow path network. A collaborative architecture of AGV agent and its different modules are also presented. Three complex experimental scenarios are simulated to test the robustness of the proposed approach. It is shown that the proposed agent-based controller is capable of generating optimal, collision- and deadlock-free path with less computational efforts.