Research Of Path Planning For Mobile Robots Based On Particle Swarm Optimization

The Particle swarm optimization (PSO) algorithm in robot pathplanning is researched in this paper. Take full advantage of the optimizeperformance of particle swarm optimization algorithm in the researchprocess, and applied to the optimization of robot path planning to improveits performance. The main research content and achievements include:Firstly, the background and significance of the topic selection isdescribed, the development and research survey of the robot at home andabroad are introduced. Several common global and local path planningmethod are introduced and these methods are analyzed and compared to laid an important foundation of the thesis research work.Secondly, the origin and the research hotspot of Particle SwarmOptimization algorithm are described. The biological mechanisms, theiterative process, the algorithm characteristics and convergence of standardParticle Swarm Optimization algorithm are presented. Meanwhile, someclassical improved Particle Swarm Optimization algorithm is introduced.The third, through the optimization of the mechanism of particle swarmoptimization analysis and combined with the background of the applicationof the robot path planning, a Particle Swarm Optimization algorithm basedon average particle size from the parameters is proposed. In order to solvethe precocious puberty of the particle swarm algorithm in the evolutionprocess, particle swarm is initialized according to a certain proportion whenthe average distance does not meet the threshold, maintaining the diversityof the population in the evolutionary process, increasing the global searchability of the group and improving the algorithm prone to precociouspuberty. The fourth, the whole simulation processes of the improved ParticleSwarm Optimization algorithm applying to robot path planning isintroduced in details. The experimental result analysis of this algorithm indifferent circumstances and the comparative results of the improved ParticleSwarm Optimization algorithm, the classical Particle Swarm Optimizationalgorithm and the standard Particle Swarm Optimization algorithmdemonstrated that the proposed method improved the diversity of thepopulation and avoided the population into a local optimum.