| Unmanned Aerial Vehicle(UAV)is an unmanned aerial vehicle that can be operated by a control device using radio remote control equipment or carrying the relevant program itself.At present,the application of single UAVs in the logistics end environment has become mature,but most of the single UAVs need human control and cannot adapt to the development trend of multi-UAV cooperation.In addition the new needs of dynamic distribution arising from the logistics end distribution link cannot be effectively solved for the time being.In the process of distribution,the problem of prolonged delivery time and increased delivery cost caused by poor ground transportation makes the user experience poor.To summarize the above background,this paper takes unmanned dynamic distribution at the end of logistics as the research object,learns and summarizes the current situation of research on multi-drone task allocation and formation control,and builds a dynamic distribution task allocation model by combining the business scenarios of end-of-logistics distribution,and designs and implements an unmanned dynamic distribution system at the end of logistics.Firstly,for the situation that the delivery parcels exceed the maximum load of a single UAV,a scheme of splitting and re-formation according to the affiliation relationship between the UAV and the sub-formation is proposed.Secondly,with the goal of minimizing the total time for multiple UAVs to complete the logistics end distribution tasks,a task allocation model is constructed by combining the needs of dynamic distribution at the logistics end.In order to improve the efficiency of handling the multi-UAV task allocation problem,an improved genetic algorithm with greedy algorithm is proposed,which can effectively reduce the computation time of the algorithm after optimization,and the solution accuracy is relatively good.Finally,the unmanned dynamic distribution system at the end of logistics is designed and implemented,which contains several modules such as UAV management and distribution task management.And the system’s functions are tested accordingly,and the test results basically meet the expectations.The research in this paper has positive significance to meet the new distribution demand at the end of logistics,and provides a new solution to the dynamic distribution problem at the end of logistics. |