Research On The Algorithm Of Task Assignment And Path Planning For AGV In Intelligent Warehouse Environment

With the rise of e-commerce,express business increased sharply.The traditional warehouse can not satisfy the enormous demand of express goods by manual work.It is urgent to improve the efficiency of warehouse with intelligent logistics device.As one of the key intelligent device in warehouse operation system,Automated Guided Vehicles(AGV)plays an unsubstitutable role in reducing logistics cost and improving logistics efficiency.The algorithm of AGV task assignment and AGV path planning is the core technology of AGV operation.Reasonable algorithm design can solve the practical problems such as long task allocation time and long driving distance.Modern e-commerce storage orders are large,the goods sorting accuracy is high,and the delivery time is high,so it is critical to research the AGV task allocation and path planning algorithm with high accuracy and can effectively reduce the operation time to enhance the warehouse operation efficiency.In this thesis,the matter and algorithm of multi-AGV task allocation in warehouse environment are studied.On the basis of task assignment,the route planning problem of multi AGV is transformed into the single AGV.Through improved algorithms,the task assignment and path planning of warehouse AGV are optimized,and the effectiveness of the improved algorithms is verified by MATLAB.The research in this thesis can effectively improve the efficiency of storage operation,and has certain practical significance and theoretical reference value for future research.The main contents of this thesis are as follows:In view of the long task assignment time of AGV,this thesis firstly describes the tasks of warehouse AGV.Then,considering the constraints of AGV power and task execution time,the task assignment model of warehouse AGV is established with the goal of minimum total travel and minimum task completion time of AGV Finally,a Consensus-Based Bundle Algorithm(CBBA)is proposed.The improved algorithm constructs association according to the location of tasks,and groups the tasks with higher correlation degree according to AGV number to form grouping strategy and introduce CBBA task assignment.The bidding function of AGV in CBBA is simplified from the profit to the path cost.Using MATLAB to simulate the CBBA before and after the improvement,the result shows that compared with CBBA,the total distance of the improved CBBA is reduced by 27.8%,the total time to accomplish tasks decreased by35.16%,the number of optimizations to reach the minimum objective function is also reduced from 47 to 2.The improved algorithm can effectively reduce the total travel distance and the task completion times,decrease the number of communication,so as to enhance the efficiency of task assignment.In this thesis,aiming at the ant colony algorithm is easy to fall into local optimization for AGV path planning,the grid method is used to establish the environment model,and the AGV driving speed is determined under the condition of sufficient power supply.Taking the shortest driving path as the objective function,an improved ant colony algorithm based on genetic mechanism is proposed.Genetic algorithm is used to search in the early stage of the improved algorithm.Mutation operator is added to avoid the interference of poor path and to accelerate the algorithm convergence.Directional and random node search is used to change the initial population generation mode and increase the diversity of initial population.The length of path is shortened by adding deletion operator.The initial optimal path of the ant colony is optimized according to the genetic search result and the pheromone assignment is updated to prevent the path from falling into local optimization.Thus the search efficiency is improved effectively.Use MATLAB in 10* 10,20*20,30*30random obstacle map environment to simulate the ant colony algorithm and improved ant colony algorithm.The comparison results show that the improved ant colony algorithm can shorten the time to find the optimal path,reduce the number of iterations,in a larger map environment can better reduce the shortest path length of AGV.The improved algorithm can solve the problem of long driving path caused by the algorithm easily falling into local optimum.Thus,the improved algorithm can solve the problem of long driving path caused by ant colony algorithm easily falling into local optimum.