| With the rapid development of society and the advancement of science and technology,intelligent unmanned equipment represented by drones has played an important role in the military and civilian fields,but the current autonomous flight capabilities of drones are still insufficient,resulting in air crashes.The accident has caused many scholars to pay attention to the research of path planning.Path planning is an important technology for drones to fly autonomously,and it is the key to ensuring that drones complete their missions smoothly.In order to ensure that the fixed-wing UAV in the complex airspace environment can successfully complete the designated tasks,this thesis studies the global path planning method based on the improved particle swarm algorithm and the local path planning method based on the improved artificial potential field method,and proposes a combination of the above two The advantages of the algorithm can quickly plan a reasonable and feasible path for the UAV.The following studies were completed during the project development process:1)In view of the fact that the particle swarm algorithm used in path planning is easy to fall into the local optimum and the convergence speed is slow in the later stage,the adaptive Cauchy Mutation Particle Swarm Algorithm(ACMPSO)is proposed.The algorithm is based on the improvement of the particle swarm optimization algorithm PSO(Particle Swarm Optimization).Firstly,with the help of exponential inertia weight adjustment strategy,the global search ability and local search ability of particles are dynamically adjusted;secondly,Cauchy mutation is introduced,and the change is set.The asynchronous length is in exponential form,so that it decreases rapidly with the increase of the number of iterations,so that the algorithm generates a larger disturbance in the early stage to help the particles jump out of the local optimum,and uses a smaller step size in the later stage to accelerate convergence.2)Based on the modeling of the 3D environment of UAV flight and the fitness function of particle swarm algorithm,a 3D environment model and fitness function model are established,and the ACMPSO algorithm is verified by the experimental simulation of UAV 3D path planning The high stability,the rapidness of convergence speed and the superiority of the quality of the generated track.3)Based on the analysis of the artificial potential field method,aiming at the unreachable problem of the algorithm,the relative distance between the UAV and the target point is introduced into the repulsive potential field function,so that when the UAV is close to the target point,The repulsion of obstacles to the UAV gradually decreases,and finally the UAV reaches the target point,which solves the problem of unreachable target.A target segmentation method is proposed to solve the problem of the algorithm falling into a local minimum.This method combines the virtual target The point method is extended from a two-dimensional plane to a three-dimensional space.After a reasonable sub-target point is generated,it is ensured that the UAV can escape the current local minimum point and reach the sub-target point.After the UAV reaches the sub-target point,it will go to the original point.Target.The simulation experiment results in a complex obstacle environment verify that the improved artificial potential field method can overcome the shortcomings of the traditional artificial potential field method,and the improved local path planning algorithm can make the UAV fly safely in a complex three-dimensional environment.4)Finally,a fusion algorithm based on improved artificial potential field method and improved particle swarm algorithm is proposed.In view of the good local planning effect of the artificial potential field method,and the good global planning effect of the particle swarm algorithm,the fusion algorithm can make the advantages of the two complementary,which is beneficial to the UAV to efficiently and quickly plan the path in the complex three-dimensional space. |
PDF Full Text Request