A Congestion-Predicting Path Planning Algorithm for Narrow Lanes in Automated Warehouses

With the gradual development of intelligent logistics and multi-AGV (Automatic Guided Vehicle) systems, the application of automated warehouses is becoming more and more extensive. Among them, the path planning of AGVs is often a huge challenge. Under the specific constraints of road layout, the path planning methods described in the existing literature are often not good at avoiding congestion. In this paper, according to the limits of narrow lanes, we propose a path planning algorithm for predicting congestion. This algorithm is an improvement of the A-star algorithm, which introduces the current congestion cost and future congestion cost, and can plan the path to avoid the congested area, reduce the probability of runtime conflicts. Through simulation comparison, it is verified that the proposed algorithm has a good application prospect in alleviating traffic congestion.

I. Introduction

In recent years, factories have gradually begun to use automated equipment to replace manual operations. Automatic guided vehicles (AGV) are one of the high-efficiency transportation tools, which can replace manual operations to realize full automation of the transportation process. It has important theoretical value and practical significance to improve the operation level and improve the efficiency of goods production and transportation [1]. Environmental modeling, AGV path planning, and multi AGV conflict free operations are its key technologies.