Research On Scheduling Algorithm And Design Of AGV

Automated Guided Vehicle is the core of the automated sorting warehouse.The picking efficiency of the automated sorting warehouse is the standard for testing the sorting ability of logistics enterprises.Intelligent and efficient AGV equipment and high-performance scheduling algorithms can effectively improve the picking efficiency of the automatic sorting warehouse and reduce the cost of sorting.In this paper,by designing the mechanical structure and on-board control system of the AGV,the AGV sorting capacity is enhanced,and the AGV path planning algorithm and scheduling algorithm are improved at the same time.The supporting upper computer system is developed to improve the sorting capacity of the automatic sorting warehouse.The main research contents are as follows:Firstly,the functional requirements of AGV will be summarized to choose the driving method and guiding method,and it also can determine the AGV design plan and specific parameters.According to the design plan of AGV,the mechanical structure of AGV is modularized to driving components,jacking components and slewing components,and the type of the motors and sensors in each component will be selected.The vehicle control system is designed according to design plan.STM32F407VGT6 is selected as the main control chip,and electrical components are selected.for motor drive module,navigation module,communication module and obstacle avoidance module,corresponding circuit diagrams are also designed.Secondly,aiming at the shortcomings of ant colony algorithm in path planning,such as slow optimization efficiency and poor path feasibility,this paper uses the grid method to construct the electronic map model of the sorting warehouse,and proposes an improved ant colony algorithm.The algorithm first speeds up the early convergence speed by adjusting the initial pheromone content of different regions in the grid map.Then it introduces an optimization strategy and pheromone negative feedback mechanism into the node transfer formula to improve the algorithm’s optimization ability.Finnally,the reward and punishment mechanism is used to improve pheromone update method,and the obtained path is crossed to improve the overall performance of the algorithm.Two simulation experiments are also carried out to verify the excellent convergence speed and optimization ability of the improved ant colony algorithm in solving the single AGV path planning problem.Then,aiming at the multi-AGV task scheduling problem,this paper establishes a corresponding mathematical model with the total cost of driving as the optimization objective,and proposes an improved sparrow search algorithm in view of the limitations of each algorithm in solving this problem..The algorithm uses the way of chaotic initialization to obtain a discretely distributed initial population,and it uses different crossover or neighborhood search methods for different types of sparrows as a local optimization strategy to improve the optimization efficiency of the algorithm.In order to verify the performance of the improved algorithm,three sets of calculation examples are selected to design comparative experiments.The experimental results prove that the improved sparrow search algorithm has reliable global stability and solution quality when solving multiple AGV task scheduling problems.Finnally,the host computer control software is designed for the automated storage system.This part will introduce the specific functional modules of the host computer system in detail and build a single AGV control platform and a multiple AGV dispatching experiment platform.The single AGV control experiment platform will test the AGV experimental prototype and the basic functions of the host computer system by controling AGV to complete the start-stop,turning,and automatic path finding commands and testing whether the AGV can accurately display the map information and AGV information.The multi-AGV scheduling experiment platform will verify the overall performance and conflict resolution strategy.