Virtual grid layout with direction constraints for autonomous mobile robot routing performance improvement

Recently, autonomous mobile robots (AMRs) have been receiving considerable attention. Unlike traditional automated guided vehicles, AMRs are highly flexible because they can move in any path without guide rails. However, AMRs in narrow regions require very careful control because the higher the number of autonomous mobile robots, the greater the probability of collisions and deadlocks. This study proposes a design methodology for a virtual grid layout with direction constraints for efficient AMR routing in a factory with limited workspace. To demonstrate the effectiveness of the proposed methodology, several simulation models of two factories were constructed using FlexSim, which supports modules for autonomous mobile robots simulation. AMR routing using the proposed virtual grid layout with direction constraints performed considerably better than conventional guided path-based routing as well as AMR routing with no restriction with respect to average lead-time and number of deadlocks. In the sample layout experiments, the proposed method demonstrated a remarkable 39.5% reduction in delivery time compared to the guided path-based routing. Furthermore, unlike the unrestricted AMR routing, the proposed method did not experience any deadlocks.