Multi-Objective 3D Path Planning For UAV Under Mobile Threats

With the increasing application of UAV technology in intelligent warfare,the autonomous flight of UAV in complex environments has gradually become a hot research topic.In order to realize autonomous flight in complex environments,UAVs usually must have excellent autonomous navigation capabilities.As a key link in the autonomous navigation process of UAV,the three-dimensional path planning of UAV not only has high space complexity,but also needs to be optimized for multiple goals.This thesis focuses on the problem of multi-target 3D path planning for UAV in combat scenarios,and proposes a 3D path planning method that combines offline planning and online planning,so as to obtain the optimal path that can meet multiple optimization goals at the same time.The UAV can safely avoid various dynamic and static threats in the scene and quickly reach the target point.The main content of the thesis includes:(1)Modeling of 3D path planning in adversarial environment.In view of the various dynamic threats and static threats that may exist in the combat scene,we establish the three-dimensional threat probability models;at the same time,the physical performance constraint models of the UAV are established in the thesis;and we establish the comprehensive cost model of the path by analyzing the optimization target of the three-dimensional path,which provides a theoretical basis for the evaluation of the merits of the path.(2)Offline 3D path planning for UAV.We predict the patrol trajectory of the potential unknown dynamic threats in combat scenarios,and generate a rasterized patrol probability model of dynamic threats,which can improve the safety of offline path planning;by combining with the probability model of dynamic threats,we realize offline 3D path planning of UAV based on genetic algorithm to obtain a pre-planned path that can satisfy the multi-objective constraints.(3)Online 3D path planning for UAV.Aiming at the sudden dynamic threat encountered during the flight of UAV,a real-time trajectory prediction method of the dynamic threat based on Rapid-exploration Random Tree algorithm is proposed,which realizes real-time and efficient prediction of the sudden dynamic threat trajectory;in order to solve the problem of slow convergence speed and poor local search ability in the later stage of the standard PSO algorithm,the hierarchical idea of combining global search and local search is introduced into the algorithm,and an online 3D path planning method based on the improved PSO algorithm is proposed.By combining with the real-time trajectory of the sudden dynamic threat,UAV can safely and efficiently avoid the sudden dynamic threat.The results of comparative experiments with existing planning algorithms demonstrate that the results verify that the three-dimensional path planning algorithm proposed in this thesis can ensure that UAV can safely avoid dynamic and static threats in combat scenarios,and can make the final path meet the multi-objective constraints such as accuracy and real-time.