Abstract In an automated stereoscopic warehouse, the efficiency of the Rail Guided Vehicle (RGV) is the bottleneck. This paper proposes a foresight stepping model to optimize the intelligent RGV scheduling scheme. We incorporate the chaotic particle swarm optimization algorithm into the model and design the mechanism of multi-step processing. The machine optimization is used to compare the optimal alignment effect of the Back Propagation (BP) network algorithm and GradientBoostingDecisionTree (GBDT) algorithm. The real-life system test is performed by simulation. The simulation results show that the GBDT-foresight stepping model is superior to the traditional models in terms of complexity, reliability and accuracy.

中文翻译：

---

智能轨道车辆动态调度优化策略

摘要 在自动化立体仓库中，轨道导引车（RGV）的效率是瓶颈。本文提出了一种前瞻步进模型来优化智能RGV调度方案。我们将混沌粒子群优化算法纳入模型并设计了多步处理机制。机器优化用于比较反向传播（BP）网络算法和GradientBoostingDecisionTree（GBDT）算法的最优对齐效果。实际系统测试是通过模拟进行的。仿真结果表明，GBDT-foresight步进模型在复杂性、可靠性和准确性方面均优于传统模型。