Application of UAV path planning based on parameter optimization GA-PSO fusion algorithm

Abstract

Abstract Aiming at the complexity of the unmanned aerial vehicle (UAV) path planning problem and the great influence of genetic algorithm parameters on the stability of the results, a fusion algorithm based on parameter optimization is proposed in this paper. In the iterative process, the GA-PSO fusion algorithm uses particle swarm optimization algorithm to search the optimal value of crossover rate and mutation rate in genetic algorithm, which makes the algorithm convergence speed is fast and search ability is strong. In addition, the core part of the algorithm fusion framework is realized by introducing the optimal adaptive value of the population after crossover and mutation as the adaptive value of the parameter particle. Finally, we design two groups of experiments and compare the proposed fusion algorithm with the classical particle swarm optimization algorithm (PSO), ant colony algorithm (ACA), genetic algorithm (GA), artificial fish swarm algorithm (AFSA), Wolf pack algorithm (WPA), artificial bee colony algorithm (ABC) and the improved algorithm through experimental simulation. The experimental results show that: In general, the path planned by the GA-PSO fusion algorithm in this paper is 10% shorter than that planned by other algorithms on average. Simulation results also show that the convergence speed of the fusion algorithm in this paper is faster, and the final search path is smoother.