Research On Warehouse Multi-AGV Dynamic Scheduling System

As an indispensable core component of the warehousing and logistics link in the modern industrial system,the intelligent warehousing system is developing from the mechanical automation to the intelligent level of the robot,especially in recent years,the booming of emerging industries such as callers and express delivery has greatly improved.It promotes the intelligent and information construction process of the intelligent storage system.The warehouse-type multi-AGV is widely used,which can replace the efficient selection of the picking personnel and realize the effective connection of the operation links.How to play the multi-AGV collaborative operation ability and improve system efficiency is the core and key research content in this field.Based on the online task scheduling model and the general warehouse space environment model,this paper conducts an in-depth study on the spatial modeling,path planning,task assignment and scheduling of intelligent warehouse systems,in order to realize an efficient warehouse-based multi-AGV dynamic scheduling system.This paper studies from the following aspects:Firstly,the structure of the intelligent storage AGV scheduling system is studied,and the functions that the system needs to implement are designed.By analyzing the layout features and implementation functions of the general storage space,the storage space mapping model of the two-layer control system is constructed,which realizes the positioning,navigation and visual management control functions of the AGV.Then,based on the established general storage space model,based on the analysis of the performance and range of the single AGV path planning algorithm,the improved values of different algorithms of the multi-AGV dynamic system are determined.Aiming at the complex relationship between AGVs,an improved algorithm for automatically adjusting the order of AGV priorities is proposed.The algorithm uses traffic rules to resolve time and space conflicts.In order to further improve traffic congestion and improve system dynamics and system flexibility,a multi-step forward-looking heuristic algorithm is proposed.The algorithm iteratively plans the optimal path to prevent congestion,resolve conflicts,and reduce planned losses in advance.Then,the task allocation mode,job flow and task scheduling constraints of the multi-AGV scheduling system are analyzed.Under the premise of satisfying the system operation characteristics,the main goal of minimizing the maximum task completion time is proposed.And the dynamic task scheduling strategy of real-time planning performance,establish the mathematical model of composite index from different angles,solve the dynamic task scheduling problem under different working conditions,and realize the optimal task scheduling of the overall system performance.Finally,combined with the theoretical research results of warehouse-based multi-AGV dynamic path planning algorithm and scheduling strategy,based on QT graphical program development framework,a warehouse-based multi-AGV scheduling system PC management software including map management,task management and system management is realized.And verify the application of the simulated working conditions,and propose a guiding working mode scheme under different working conditions.