Research On Logistic UAV Swarm Routing With Simultaneously Pickup And Delivery Based On Improved Teaching-Learning-Based Optimization

Logistics is an important part of China s national economy.Transportation cost accouts for more than 50%of the total logistics cost.Using Express Unmanned aerial vehicle(UAV)for distribution and transportation can effectively reduce the cost of distribution.A a reasonable and efficient distribution scheme for express UAV swarm s delivery and pickup can further save transportation costs.This problem can be regarded as vehicle routing problem(VRP).With the need of practical logistics distribution,Express UAV s distribution and transportation is combined with vehicle routing problem with simultaneous delivery and pickup and time window(VRPSPDTW),so that the total cost of logistics ransportation can be effectively reduced.Therefore,in order to solve the route planning problem of express UAV swarm,this paper propose an algorithm called Improved Teaching-Learning-Based Optimization algorithm(HTLBO).With the benchmark data,the HTLBO algorithm is compared with the existing methods.The effectiveness of the proposed algorithm is verified.The HTLBO is also tested by the customers' data in two sampling areas,which proves the practicability of the algorithm.The main contents of this paper are as follows:(1)This paper firtsly introduces the advantages and necessity of using express UAV as the means of distribution and transportation.The problem of VRPSPDTW is discussed by combining the route planning with express UAVswarm s simultaneously pickup and delivery.Then the paper introduces the state of art of VRPSPDTW and express UAV's routing.(2)Based on the mathematical programming model of classical vehicle routing problem,this paper introduces the composition,parameters,optimization objectives and constraints of VRPSPDTW problem respectively.We also present the mathematical programming model of VRPSPDTW problem in this part.(3)Aiming at the VRPSPDTW problem with express UAV swarm s distribution and transportation,the proposed HTLBO algorithm in this paper is improved from the aspects of solution coding,evaluation and feasibility analysis of a solution,population initialization,solution transformation,solution updating.In addition,the main innovation of this paper is in that it is the first time to combine the HTLBO algorithm with the vehicle routing problem and we propose a new coding method.On the basis of the original algorithm,it innovatively uses a student self-study phase and a teacher self-learning phase.These two phases are based on a depart operation and an improved variable neighborhood search algorithm respectively.The Simulation results proves the effectiveness of our proposed HTLBO algorithm.