| The Automated Guided Vehicle is a new type of robot that integrates intelligent technologies such as external intelligent perception and automatic execution.It can not only quickly and efficiently accomplish the tasks of unmanned transportation and cargo handling,but also be an important part of logistics and industrial automation.The use of reasonable scheduling and planning algorithms can improve material turnover and reduce logistics costs.This article focuses on multi-AGV scheduling for enterprise internal logistics.By optimizing the task queue,the transportation distance is minimized,transportation costs are further reduced so that efficiency is improved.simultaneously,considering the internal data security of the enterprise,the system of user management is realized based on the auxiliary storage of the industrial block chain,which further improves the security of the scheduling system account and can prevent data leakage to a certain extent.Research on the internal logistics business of the enterprise,the function design of the scheduling system has been completed,and the scheduling and path planning algorithms of multiple AGVs have been studied in depth,which can effectively save the cost of internal logistics of the enterprise and also effectively improve the efficiency of material transportation.The main research contents are as follows.By analyzing the enterprise's internal logistics business requirements,consider the convenience of users and principle of modularization.The workflow of the entire system is planned,and a three-tier technical architecture is adopted to design and improve the multiAGV scheduling system.Based on the factory model,the AGV driver is designed to realize different types of AGVs running and participating in the scheduling in the scheduling system at the same time.In view of the optimal principle of AGV utilization and path distance,the task model and scheduling strategy based on multi-priority are realized.Tasks are divided into different priorities,when a task is created,the scheduling system first calculates the size of the task priority and then puts it into the corresponding task queue to wait for operation.The priority of the same level is scheduled according to the first creation first execution rule.Aiming at the time-consuming problem of AGV steering,the A* algorithm is further improved by analyzing and comparing the three evaluation functions of A* algorithm.Firstly,the proportion of the Manhattan evaluation function is increased and the steering weight is added to the heuristic function to ensure the path of the AGV when performing tasks in order to keep the vehicle to go straight,reduce the number of turns and improve transportation efficiency.Secondly,consider path conflicts and deadlocks in static planning,design a time window of refined calculation and propose a static time window strategy based on task priority are used to reduce the probability of conflicts and deadlocks.Finally,for the runtime failures and conflicts,real-time monitoring and dynamic adjustment strategies are proposed.The system monitors the whole process in real-time during the execution of tasks,predicts and avoids conflicts to improve system stability.Finally,an experiment platform for multiple AGVs was built.First,a single AGV scheduling experiment was conducted to verify that the scheduling system function and path algorithm can achieve the expected results.secondly,the multi-subtask random assignment to multiple AGV experiments was used to verify that the scheduling system is uniform in a multi-task environment,which can map the shortest and least turning paths,and multiple AGVs can efficiently complete the transportation tasks without conflicts.It further verifies that the time window strategy can effectively avoid path conflicts and deadlocks.Upcoming conflicts are predicted and avoided,the effectiveness of real-time monitoring and dynamic adjustment strategies are tested based on the multi-AGV operating environment. |
PDF Full Text Request