• CN:11-2187/TH
  • ISSN:0577-6686

机械工程学报 ›› 2021, Vol. 57 ›› Issue (10): 245-256.doi: 10.3901/JME.2021.10.245

• 交叉与前沿 • 上一篇    下一篇

扫码分享

密集仓储环境下多AGV/RGV调度方法研究

周亚勤, 汪俊亮, 吕志军, 项前, 丁扬, 张洁   

  1. 东华大学机械工程学院 上海 201620
  • 收稿日期:2020-06-30 修回日期:2021-02-23 出版日期:2021-07-23 发布日期:2021-07-23
  • 通讯作者: 周亚勤(通信作者),女,1977年出生,博士,副教授。主要研究方向为智能车间生产调度建模、优化调度算法。E-mail:zhouyaqin@dhu.edu.cn
  • 作者简介:汪俊亮,男,1991年出生,博士,副研究员。主要研究方向为工业大数据分析、智能制造系统等。E-mail:junliangwang@dhu.edu.cn;吕志军,男,1967年出生,博士,副教授。主要研究方向为智能检测、轻量化设计、密集仓储。E-mail:lvzj@dhu.edu.cn;项前,男,1973年出生,博士,副教授。主要研究方向为数字化制造、智慧物流、计算智能及应用。E-mail:xqsir@dhu.edu.cn;丁扬,男,1993年出生。主要研究方向为AGV小车调度。E-mail:dingyang@dhu.edu.cn;张洁,女,1963年出生,博士,教授,博士研究生导师。主要研究方向为智能制造系统与大数据技术、先进制造系统智能优化与调度方法、制造业信息化与智能化关键技术等。E-mail:mezhangjie@dhu.edu.cn
  • 基金资助:
    国家自然科学基金(51905091)、上海市科技计划(20Dż2251400)和上海市工程技术研究中心能力提升计划(17DZ2283800)资助项目。

Research on Multi-AGV/RGV Scheduling Method in Intensive Storage Environment

ZHOU Yaqin, WANG Junliang, Lü Zhijun, XIANG Qian, DING Yang, ZHANG Jie   

  1. School of Mechanical Engineering, Donghua University, Shanghai 201620
  • Received:2020-06-30 Revised:2021-02-23 Online:2021-07-23 Published:2021-07-23

摘要: 针对密集仓储环境下,出库作业时,有轨车(Rail guided vehicle,RGV)在不同货架间运送货物的换乘需要借助穿梭车(Automated guided vehicle,AGV)实现,入库作业时,货物先由穿梭车从输送带送达货架口,再由有轨车完成入库操作等特征,构建密集仓储环境下考虑多出入库任务的多AGV/RGV作业调度模型,包括穿梭车任务分配模型、协同有轨车选择模型和出入库完工时间数学模型。为实现密集仓储环境下的多AGV/RGV调度,提出适应不同出入库货位分布的穿梭车任务分配规则,实现考虑执行任务均衡的穿梭车任务分配;利用遗传算法实现多AGV/RGV出入库协同调度,对遗传算法关键解码算子进行详细设计,解码确定各穿梭车与有轨车执行出入库任务的顺序、任务的起始时间和结束时间,使得所有出入库任务的总完工时间最短。最后,通过某物流仓储企业实际案例进行测试,测试结果表明,提出的启发式规则能实现穿梭车任务的均衡分配,基于遗传算法的协同调度方法能有效地产生多AGV/RGV协同调度方案,减少出入库作业总时间,提高了仓储作业整体效率。

关键词: AGV/RGV协同调度, 密集仓储, 遗传算法

Abstract: For intensive warehouse storage operations, the automated guided vehicle(AGV) is needed to transfer the rail guided vehicle(RGV) between the shelves to transports the goods. When the storage is in operation, the AGV delivers the goods from the conveyor belt to the shelf, then RGV cooperates to complete the storage operations. It is important to construct a cooperative scheduling model for AGV/RGV considering the multiple warehousing tasks in an intensive storage environment, which includes AGV task assignment model, coordinated RGV selection model and completion time mathematical model. In order to realize multi-AGV/RGV scheduling problem, the task allocation rules of shuttle vehicles adapting to the distribution of different inbound and outbound cargo spaces are proposed with the objective of balancing the AGV tasks. Genetic algorithm(GA) is proposed to solve the multi-AGV/RGV inbound and outbound cooperation scheduling. The key decoding operator is designed in detail to determine the task sequence, task start and end time of AGV and RGV to perform the inbound and outbound tasks, so that the total completion time of all inbound and outbound operations is the shortest. Finally, the actual case of a logistics warehousing enterprise is tested. The case study demonstrates that the proposed heuristic rules can achieve the balanced distribution of AGV tasks, and the GA-based cooperative scheduling method can effectively generate multi-AGV/RGV coordinated scheduling scheme in terms of reducing the total time of warehousing operations and improving the overall efficiency of warehousing system.

Key words: AGV/RGV collaborative scheduling, intensive storage system, genetic algorithm

中图分类号: