面向绿色再制造系统的AGV路径规划研究

Study on AGV path planning for Green Remanufacturing System

摘要:

为了解决绿色再制造系统中的自动导引运输车（AGV）路径规划问题的问题，提出一种粒子群遗传融合的AGV全局路径优化的自适应算法，该方法不仅集成了遗传算法(GA)和粒子群算法(PSO)二者的优点，而且为了改善传统PSO-GA融合算法迭代前期寻优速度慢的问题，引入自适应惯性权重；为了提高算法进入迭代后期的收敛精度，提出了一种双重交叉变异策略,使得改进的PSO-GA融合算法比传统的PSO-GA融合算法搜索能力更强，进化速度更快，收敛精度更高，为了验证改进后算法的优越性，采用栅格法模拟自动导引运输车运行环境并通过MATLAB对标准粒子群、遗传、传统的PSO-GA融合、改进PSO-GA融合四种算法解决路径优化问题进行试验对比，结果证明了改进后的PSO-GA算法的可行性和有效性。

Abstract:

In order to solve the problem of automatic guided vehicle (AGV) path planning in green remanufacturing system, an adaptive algorithm for global path optimization of AGV based on particle swarm optimization (PSO) is proposed. This method not only integrates the advantages of genetic algorithm (GA) and particle swarm optimization (PSO), but also improves the slow search speed of traditional fusion algorithm in the early iteration stage. In order to improve the convergence accuracy of the algorithm in the later iteration stage, a dual crossover mutation strategy is proposed. The improved PSO-GA fusion algorithm has stronger search ability, faster evolution speed and higher convergence precision than the traditional PSO-GA fusion algorithm. In order to verify the superiority of the improved algorithm, the grid method is used to simulate the running environment of the auto-guided transport vehicle, and the four algorithms of standard particle swarm optimization, genetic algorithm, traditional PSO-GA fusion and improved PSO-GA fusion are solved by MATLAB. The experimental results show that the improved PSO-GA algorithm is feasible and effective.