Research On Path Planning Of UAV Swarm With No Fly Zone

With the rapid development of the logistics industry,new means of transportation and modes of transportation emerge in an endless stream.Choosing suitable means of transportation and designing reasonable means of transportation can effectively improve the quality of logistics distribution.UAV is light,flexible and controllable,Express-UAVs have become a feasible solution for logistics.Due to the strict aviation control in China,the flight route of UAV needs to avoid the no fly zone.Therefore,it is of great value and significance to plan reasonable obstacle avoidance route and design efficient pathing for UAV.In this paper,the path planning of logistics drone swarms with no-fly zone constraints is the main research content,and a hybrid particle swarm algorithm with obstacle avoidance function is designed to solve this problem.The main objective is to minimize the number of UAVs and the secondary objective is to minimize the total power consumption of UAVs,two coding methods are used to express the solution,a new route splitting method for battery capacity constraint is put forward to split the solution.Particle swarm optimization algorithm is used for global search,and variable neighborhood descent algorithm is used to strengthen local search,for the global optimal solution found in each iteration,the improved A-star algorithm is used to avoid obstacles and update the path.Based on the standard test cases,the experimental results of this algorithm are compared with those of particle swarm optimization algorithm,cultural gene algorithm and iterative local search algorithm,and the effectiveness of this algorithm is verified.In order to get closer to the actual scene,improve equipment utilization and reduce costs,this paper further studies the path planning problem of logistics UAV group with persistent delivery schedules.We proposed a hybrid adaptive large neighborhood search algorithm to solve the problem which a three-layer coding method is used to represent the solution,and uses a time-based splitting algorithm to allocate transportation tasks for UAVs.To further improve the quality of the solution,a new destruction recombination method is proposed,which combines with the variable neighborhood desce nt algorithm to accelerate the convergence of the solution.Through experiments,the proposed algorithm has shown better performance in terms of effectiveness and stability when compared with the improved hybrid particle swarm optimization algorithm and iterative local search algorithm.