An Improved Particle Swarm Algorithm for UAV 3D Path Planning

For the problem of three-dimensional path planning for UAVs, an improved PSO(Particle Swarm Optimization) is proposed by improving the existing model. For the model, the Analytic Hierarchy Process is used to determine the weight parameters for the objective function. In terms of algorithm, firstly, the population will be initialized at a higher altitude range pair to increase the number of effective particles in the pre-population; secondly, for the drawback of particle swarm algorithm that is easy to fall into local optimum, the inertia parameter is processed by linear varying, and an additional population is set up to enhance the exploration and development in the full cycle of the algorithm; moreover, in order to carry out the exploration and development in a better way, a real-time updating elite population is proposed which can record multiple optimal particles in the environment. The experimental results show that compared with the original algorithm, the IPSO(Improved Particle Swarm Optimization) improves the convergence speed and optimization performance.