粒子群优化
算法
渡线
遗传算法
惯性
趋同(经济学)
运动规划
计算机科学
多群优化
机器人
数学优化
工程类
人工智能
数学
物理
经典力学
经济
经济增长
作者
X. Rong Li,Bangsen Tian,S. C. Hou,Xinxin Li,Li Yang,Chong Liu,Jingmin Li
出处
期刊:Electronics
[MDPI AG]
日期:2023-07-31
卷期号:12 (15): 3289-3289
标识
DOI:10.3390/electronics12153289
摘要
To address the problem of cooperative work among right-angle coordinate robots in spacecraft structural plate mount tasks, an improved particle swarm optimization (PSO) algorithm was proposed to assign paths to three robots in a surface-mounted technology (SMT) machine. First, the optimization objective of path planning was established by analyzing the working process of the SMT machine. Then, the inertia weight update strategy was designed to overcome the early convergence of the traditional PSO algorithm, and the learning factor of each particle was calculated using fuzzy control to improve the global search capability. To deal with the concentration phenomenon of particles in the iterative process, the genetic algorithm (GA) was introduced when the particles were similar. The particles were divided into elite, high-quality, or low-quality particles according to their performance. New particles were generated through selection and crossover operations to maintain the particle diversity. The performance of the proposed algorithm was verified with the simulation results, which could shorten the planning path and quicken the convergence compared to the traditional PSO or GA. For large and complex maps, the proposed algorithm shortens the path by 7.49% and 11.49% compared to traditional PSO algorithms, and by 3.98% and 4.02% compared to GA.
科研通智能强力驱动
Strongly Powered by AbleSci AI