Path Planning Of UAV Based On Intelligent Optimization Algorithm

With the information,unmanned and intelligent development of aviation technology,Unmanned Aerial Vehicles(UAVs)have increasingly advantages in the battlefield environment.Not only are they low-cost,without casualties,but they can also sustain high-intensity operations.Among them,path planning is one of the key technologies to realize autonomous and intelligent UAVs.UAV path planning is to plan one or more safe and effective paths from the starting point to the target point by one or more optimization algorithms under the con-straints of UAV performance and various threats.In most UAV path planning studies,only a single target is considered or multiple targets are converted into a single target through a weighted sum method for solution.There are still relative-ly few path planning studies that consider multiple objectives at the same time.While the former algorithm can only plan one path at a time,it is difficult to meet the diverse needs of decision makers,and it is difficult to set weights.Meanwhile,the battlefield environment is becoming more and more complex,single UAV has a single function,low fault tolerance and limited reconnaissance capability,so it can not perform complex tasks.However,multi-UAV can coop-erate with each other with higher efficiency to accomplish tasks.Therefore,the cooperation between multi-UAV has become an inevitable choice.In view of the above problems,the studies of the multi-objecive UAV path planning consider-ing multiple indicators and multi-UAV path planning considering time coordina-tion are carried out.This thesis mainly completes the following two aspects of work:(1)Aiming at the multi-objective path planning problem of UAV in com-plex environment,the multi-objective optimization model of UAV path planning is established with path length and threat degree as optimization indexes.Then,in order to find out a group of optimal paths with diverse distribution,an im-proved NSGA-II algorithm is designed.Based on the classic intelligent multi-objective optimization algorithm NSGA-II,this method introduces the addition and deletion operators to make the planned path avoid the threat area,the maxi-mum turning angle constraint to mitigate the route mutation caused by mutation operation,and a new algorithm with mixed target space and decision space in-formation Crowding distance operator improves the diversity of paths.Finally,Matlab is used for experimental simulation,and compared with NSGA-? and traditional GA algorithm,the simulation results show that the improved NSGA-?algorithm can effectively find a group of paths with good convergence and di-verse distribution,and can meet the diversified needs of decision-makers,alt-hough the running time is slightly longer.(2)The cooperative Path Planning of multi-UAV considering time coopera-tion is studied.Firstly,combined with the characteristics of multi-UAV coopera-tive problem,the constraints of multi-UAV cooperative are analyzed,and the hi-erarchical model of system and the objective function model of multi-UAV co-operative path planning are established,in which the cost of path smoothness is considered.Then an improved genetic algorithm is designed in the path planning layer to plan the path of multi-UAV.One is to improve the initialization of the population,and the other is to introduce disturbance operator,exchange operator and add delete operator into the traditional genetic algorithm.The next,the co-operation time and cooperation function of multi-UAV are described.The coop-eration cost of the whole team is considered-in the cooperation planning layer,and the cooperation path is determined.The smooth path is obtained by cubic spline interpolation method in the smooth layer.Finally,through MATLAB sim-ulation,and compared with the traditional genetic algorithm,proved the superi-ority of the improved algorithm.