Research On Task Scheduling And Path Planning Of Multi-AGV System

Nowadays,China's industrial technology is developing at a high speed,and Automated Guided Vehicle(AGV)has been widely applied in intelligent manufacturing system.In order to improve the operation efficiency of intelligent manufacturing system and reduce industrial production cost,task scheduling and material transportation path in the system need to be optimized.Therefore,this paper will discuss the task scheduling and path planning of multi-agv system from two aspects.The main research contents are as follows:First of all,as for the task scheduling problem of multi-agv system,based on the concept of Agent,this paper constructs the task scheduling work model of multi-agv system,and studies the task scheduling problem based on the method of contract network protocol.the number of bids and the range of bid selection in the traditional contract network protocol are limited by introducing the time threshold and the strategy of trust evaluation and buffer.Then,the scheduling decision algorithm based on the improved contract network protocol is constructed.in the Matlab simulation platform,for the scheduling decision algorithm,set up the normal scheduling and interference scheduling experiments,through the gantt chart comparison and analysis,verify the effectiveness of the scheduling decision algorithm based on the improved contract network protocol.Second,For the path planning of multi-agv system,the working space of AGV is established by raster map to reduce the dimension of decision variables in path planning.At the same time,the path length,path safety and path smoothness are combined into the objective function of path optimization in a weighted form.In the later stage of the traditional genetic algorithm,there is a problem that the population is prone to lose diversity.Therefore,the update mechanism of the Wolf pack algorithm is added to improve the traditional elite retention strategy,and then the selection operation is completed by combining the roulette method.Based on the similarity coefficient of population,the traditional adaptive selection strategy is improved and the probability of crossover and mutation in genetic operation is further optimized.Finally,due to the individual randomness of genetic algorithm,a single genetic algorithm is difficult to obtain the optimal solution.Therefore,a hybrid genetic algorithm is constructed by combining the improved genetic algorithm and the Wolf pack algorithm in a hierarchical form.Third,Based on the concept of finite state machine(FSM),a path optimization model for intelligent vehicles in multi-agv system is proposed.based on Matlab simulation platform,AGV path planning simulation platform is constructed.Finally,In order to verify the effectiveness of the path planning algorithm and path optimization model in this paper,the static environment path planning experiment and the dynamic environment path planning experiment were set in the simulation platform.Through the experimental results,the effectiveness of the improved genetic algorithm and path optimization model in this paper was verified.