Research On UAV Track Planning Based On Swarm Intelligence Algorithm

With the advancement of science and technology,drones have become more complete and diversified in function,and their applications have become more extensive.They have been applied and researched in the fields of power,industry,agriculture,and production optimization,and at the same time as a New high-tech weapons and equipment also have great development potential.Automatic trajectory planning is one of the key technologies to achieve autonomous flight of drones.Track planning is achieved through the core track planning algorithm.Swarm intelligence optimization algorithm is a new kind of optimization algorithm,which has a better ability to find the solution to the relatively complicated UAV track planning problem.Among them,Particle Swarm Optimization(PSO)and Artificial Bee Colony(ABC),as two typical swarm intelligence algorithms,have the advantages of fewer parameters and strong solving ability.The field of trace planning has great research value.This paper aims at the advantages and disadvantages of the standard particle swarm algorithm and the standard artificial bee colony algorithm,and makes a single algorithm improvement.The adjustment of the inertia weight is set in the early and late stages of the particle swarm algorithm iteration,and the balance between particle inertia and optimization behavior is achieved.Introducing the idea of crossover and mutation of genetic algorithm into particle swarm algorithm,combined with 3D trajectory planning of unmanned aerial vehicle,simulation verification results show that the feasibility of improved particle swarm algorithm is improved,and the convergence speed and algorithm stability are improved.The global optimal guidance term is added to the update stage of the artificial bee colony algorithm,which has stronger guidance for bee mining.Introduce a reverse probability selection method,which is performed in parallel with the conventional selection method to avoid premature algorithm convergence.Simulation experiments verify the performance of the improved algorithm in UAV trajectory planning,and both path smoothness and search capabilities are improved.At the end of the thesis,two combined improvement measures are proposed:PSO-ABC algorithm and hybrid optimization algorithm(P-ABC),and they are applied to the UAV track planning.The PSO-ABC algorithm divides the population into A group and P group,and evolves according to the ABC algorithm and the PSO algorithm,respectively.The two subgroups can share high-quality information by exchanging the learningmethods of the optimal particles.P-ABC introduces the learning of individuals in the particle swarm algorithm to the bee colony algorithm.By sharing particles,it brings better guidance to the following bees of the bee colony,and also makes the next generation evolutionary environment of the particle swarm better.Good globality and accuracy.Finally,through simulation experiments,the effectiveness and feasibility of two combined algorithms in 3D space track planning are verified.Compared with other track planning algorithms,it has shorter path length,less time consumption,and smoother paths.The advantage is that it speeds up the convergence and improves the efficiency and stability of track planning.The improved algorithm's track planning can obtain the optimal track that meets the constraint relationship,which has important reference value for the realization of autonomous flight.