Intelligent Path Planning Algorithm Of Unmanned Aerial Vehicle

The unmanned aerial vehicle(UAV)path planning must deal with many challenges in complex environment,maintain the safety of the flight path,and minimize the length of the planned path to the greatest extent feasible in order to minimize fuel consumption.The goal of UAV path planning is essentially to find the best solution to a difficult constraint problem,thus the swarm intelligence optimization algorithms,which is simpler,more effective,and easier to use than the conventional algorithms,is frequently employed to solve the problem.A high number of path-forming points are required to create feasible paths to avoid threat sources in order to assure the safety of UAVs,which expands the dimension’s size and uses more processing power.The swarm intelligence optimization algorithm’s drawbacks,such as its limited search capability,propensity to slip into local optimization,and unpredictable planning outcomes,are exposed by the increase in dimensions.For the issue,the study makes the following contributions:(1)First of all,in order to solve the challenge that the high dimension in the path planning problem brings to the swarm intelligence optimization algorithms,this paper proposes a new model called Double Layer Coding(DLC),which reasonably reduces the output dimension of the path forming point,lowers the calculation cost,and enhances the stability of the path.(2)In order to implement the newly proposed DLC path planning model,this paper improves particle swarm optimization(PSO).For PSO,this paper introduces a new strategy of rotating particles in high-dimensional space to search for targets in it.This strategy effectively improves the search ability of PSO,and addresses the shortcomings of it,which is easy to premature and fall into local optimum.The improved PSO algorithm combined with the DLC path planning model are successfully applied to path planning in two-dimensional environments.(3)In addition,the paper improves the grey wolf optimizer(GWO),introduces adaptive learning strategy based on K-neighborhood and differential "hunger-hunting strategy " into GWO to enhance its population’s diversity and global exploration.The fitness distance correlation(FDC)technology is used to balance the global exploration and local exploitation ability of GWO.The improved GWO is combined with the DLC path planning model and successfully applied to the path planning in the three-dimensional environments.A large number of simulating experiments’ results show that the DLC path planning model and the two swarm intelligence optimization algorithms based on it can always plan the best feasible paths for UAV in complex flight environments.