The Research On Path Planning Based On Particle Swarm Optimization Algorithm

Particle swarm optimization(PSO) algorithm is a new swarm intelligence optimization algorithm, as for its simple concept, less parameters, easy realization and so on. The algorithm won the favor of many researchers at home and abroad. It is widely applied in function optimization, automatic control, machine learning, engineering design and other fields nowadays. Due to its immature development, further research is required in the theoretical basis, the improved method and application. This paper mainly focuses on the improvement of the standard PSO algorithm, and the main contents include the following aspects:Firstly, this paper introduced the research status and basic theory of PSO algorithm, and includes the principle, mathematical description, algorithm steps, algorithm flow, and the convergence of the algorithm, the main parameters of the PSO algorithm are analyzed and discussed in detail.Secondly, aimed at the problem which the path planning of mobile robot based on PSO algorithm is easy to fall into local optimal value, this paper drew lessons from the flight of geese and then proposed a global path planning of mobile robot based on improved wild geese PSO algorithm. The improved wild geese PSO algorithm used chaos to initialize the geese PSO, and handled the premature particles to avoid the premature convergence in the process of the algorithm. In addition, through introducing a new adaptive inertia weight formula, the improved wild geese PSO algorithm could better balance the global search with local search. The experimental result shown that the improved wild geese PSO algorithm could better get rid of the local optimum, and search the global optimal path.Finally, in the combinatorial optimization problems of discrete space, for more complex traveling salesman, the particle diversity decreased at PSO algorithm’s later period. the cross strategy based on genetic algorithm were proposed,which could makes the excellent gene of the previous generation of particles passed on to the next generation, then through using the heuristic factor strategy to improve particle swarm optimization(PSO) algorithm, which makes the algorithm approach the global optimal value. Simulation result shown that the improved PSO algorithm has obvious superiority in TSP problem.