Article Preview
Top1. Introduction
With the development of control engineering technology, many enterprises have gradually raised the awareness of industrial intelligent automation. Rail guide vehicle dynamic scheduling problem is an important branch in the automation industry. Nowadays, the RGV is mainly applied in production scheduling workshop, logistics transportation, component assembly, and many other fields. The RGV has the characteristics of fast running, stable performance and high automation, the dynamic scheduling problem associated with RGV can effectively improve the production efficiency of modern intelligent processing, which has been widely used in various workshops and automated warehouses. The RGV can be divided into self-driven type, passive-driven type, assembly type and transport type according to the driving mode and the purpose (Martina et al., 2018; Sáez et al., 2008). At present, most enterprises choose the ring RGV, which can run multiple vehicles on the same track and greatly improve the capacity of transportation and assembly (Lee et al., 1996).
The research on the RGV dynamic scheduling is mainly based on the warehousing system (Roy et al., 2016). Lee et al. proposed the RGV scheduling strategy based on FCFS and analyzed the system operating efficiency of different quantities the RGV in automated warehousing system (Zhu et al., 2016). Sáez et al. (2008) effectively completed the multiple RGV scheduling tasks by predicting the tasks to be generated in advance, fuzzy classification algorithm was used to calculate and generate the probability of the tasks based on historical data, and then used a genetic algorithm to find a reasonable RGV scheduling path (Gagliardi et al., 2015). Zhang et al. established the genetic algorithm model and made a comprehensive analysis of the dynamic scheduling model of loading and unloading completed by the cooperation between the RGV and CNC (Ferrara et al., 2014). Wu et al. constructed a scheduling model with the shortest running time of RGV as the objective function, and then they adopted dynamic programming algorithm based on TSP to save the scheduling time of the RGV, accordingly improved the production efficiency of intelligent machining (Roy et al., 2016). All the above researches were carried out purely from the perspective of traditional algorithms, which is used to solve the scheduling problem by setting various constraints. Meanwhile, traditional algorithms are Although the scheduling results sometimes have some rationality using traditional algorithms, they are easy to fall into the local optimum, higher complexity, and more runtime.