Hybrid particle swarm optimisation approach for 3D path planning of UAV
Yi Junkai,Sun Xueying,Chu Hongyue
DOI: https://doi.org/10.1504/ijbic.2023.136105
2024-01-18
International Journal of Bio-Inspired Computation
Abstract:Path planning plays an important role in the control of unmanned aerial vehicle (UAV) and the completion of the combat mission. Particle swarm optimisation (PSO) is an efficient approach algorithm based on birds' foraging behaviour, but it is easy to fall into local optimum. In view of the shortcomings of this algorithm, a hybrid algorithm is proposed based on the traditional PSO algorithm. The bacterial foraging algorithm is introduced to calculate the environment coefficient, migration, and the idea of crossover and mutation of genetic algorithm to improve the calculation speed and ability of the algorithm. This algorithm can not only guarantee the operation speed but also jump out of the local optimal value, which improves searchability. In the experiment, the operational constraints of the UAV in the actual scene are considered to further improve the usability of the method. The simulation experiment verifies that the hybrid algorithm jumps out of the local optimal value under the constraint, ensuring the operation speed. Compared with the traditional method, the hybrid algorithm has proved its superiority in running time and navigation planning.
computer science, artificial intelligence, theory & methods