When it comes to some problems with increasing dimension

the basic particle swarm optimization（ PSO） algorithm will have an increasing possibility of getting into local solution

an improved PSO algorithm which integrates fuzzy C means（ FCM） algorithm is proposed. When classification of FCM algorithm is stopped

the roulette method is used to determine the class that every particle belongs to

the historical best position that all particles experience in PSO algorithm velocity updating formula is changed to current（ this iteration） optimal position inner class in particle. The improved algorithm is used for unmanned aerial vehicle（ UAV） 3 D path planning simulation experiment

the simulation results show that the improved algorithm significantly improves global optimizing ability of basic PSO algorithm and optimizing precision near the optimal solution. A reasonable and effective path is obtained successfully.