Research On Path Planning Simulation Of UAV

In modern warfare,unmanned aerial vehicles(UAVs)are vital in air and joint air-ground warfare.However,safe flight from the takeoff point,avoidance of all obstacles,avoidance of attack by the enemy,and finally reaching the end-point are the primary tasks of UAVs in many tactical applications and are prerequisites for performing a series of missions.Therefore,UAVs need a flight path that meets the mission requirements and requires effective path planning methods to obtain it.In addition,conventional simulation schemes for UAV tactical path planning have caused unavoidable errors due to inadequate simulation of various factors such as the environment,situational assessment,flight control,sensing data collection,and cycle tracks,while UAV path planning research conducting real-world development and testing is another costly and tedious task.To address the above issues:(1)We present an enhanced path-planning approach based on the Gray Wolf optimization algorithm(GWO)in this study.It is used to solve problems of high complexity and early convergence of the traditional algorithm.In this modified algorithm,the Lévy flight stochastic strategy and the symbiotic biological search algorithm are added to improve GWO for the two problems of sluggish convergence and easy fall into local optimum at the late stage.By comparing this method to the other three algorithms,the effectiveness and,to some extent,the superiority of the revised algorithm in this work are demonstrated.(2)For the sudden mobile threats in the battlefield environment,this thesis predicts the information of mobile threats in the next moment based on the Kalman filter method and uses the artificial potential field method to guide the UAV to move to a safe position,to achieve the purpose of local path re-planning to avoid the threats.(3)In this paper,an AirSim-based UAV simulation platform is proposed.It avoids frequent and complex real-aircraft tests of UAV-related algorithms by mirroring reality,transfers these tasks to the simulation environment,and achieves as few changes as possible when the algorithms are transplanted from the simulation environment to the real environment to meet requirements such as cost control.This study develops the UAV path planning virtual-real mapping simulation model and simulated environment based on AirSim,incorporating the major parts of the UAV physical entities,to increase simulation accuracy and make the simulation impact more realistic.The efficiency of the relevant path planning algorithm is evaluated by simulation results analysis,which offers inspiration for algorithm development.