Abstract: A novel PSO named competition particle swarm optimization has been proposed. Firstly, every particle swarm changes its velocity toward to global best and local best at each time step. Secondly, the optimal generation was updated by comparing the two homologous particles. Thirdly,compared with other algorithms such as the BPSO, LWPSO, EPSO,TVAC and used several test functions to analyze, it is demonstrated that CPSO gets better results in an accurate, stable way. Finally, competition particle swarm algorithm is applied to the path planning for UUV path-planning and it gets the better planning track.