刘星余

摘 要：近几年来，经济的迅速发展推动了电子商务的发展，物流仓库的智能化也随着电子商务的发展而逐渐普及，物流仓库中货物的分拣一直是其发展中的难题，物流仓储分拣机器人需要的复杂的物流仓库中分拣出各式各样的目标货物，物流仓库的环境问题以及大量的货物更是加大了分拣机器人的工作难度。针对物流仓储分拣机器人分拣目标的多种多样性，以及仓库的复杂性等因素的干扰，本文研究了面向物流仓储分拣机器人的多目标视觉识别与定位方法，介绍了分拣机器人的目标识别与定位系统结构，以及分拣目标的提取、识别与定位，并进行了实验验证，对于物流仓储中分拣机器人的研究及应用有重要意义。

关键词：物流仓储;分拣机器人;多目标;识别与定位

中图分类号：TP242                          文献标识码：A                         文章编号：1001-5922（2021）07-0109-04

Research on Multi-Target Visual Recognition and Positioning Method for Logistics Warehouse Sorting Robot

Liu Xingyu

（Sichuan Vocational College of Information Technology， Guangyuan 628017， China）

Abstract：In recent years， the rapid development of economy has promoted the development of e-commerce. The intelligentization of logistics warehouses has gradually become popular with the development of e-commerce. The sorting of goods in logistics warehouses has always been a problem in its development. Logistics storage sorting Robots need to sort out a variety of target goods in the complex logistics warehouse. The environmental problems of the logistics warehouse and a large number of goods make the task of sorting robots more difficult. In view of the interference of the diversity of sorting targets ， as well as the complexity of warehouses and other factors， the paper studies the multi-target visual recognition and positioning method for logistics warehouse sorting robots， and introduces the structure of multi target recognition and positioning system of the sorting robots， as well as the extraction， recognition and positioning of sorting target， and experimental verification have been carried out， which are of great significance to the research and application of sorting robots in logistics warehousing.

Key words：logistics warehousing; sorting robot; multi-target; recognition and positioning

0 引言

物流倉储分拣机器人属于工业机器人的范畴，具体是指在物流仓库中，通过指令的接收或者根据预先设定的任务目标，自动进行货物的分拣、移动、搬运等操作的机器装置。物流仓储分拣机器人是智慧物流的重要组成，符合时代发展的潮流和要求，其发展对于电商物流中过分依赖人共、业务高峰期分拣能力较差等瓶颈问题有重要的意义。物流仓储机器人根据应用场景的不同可以分为五大类，分别为AGV机器人、码垛机器人、分拣机器人、AMR机器人、RGV穿梭车，本文中主要对分拣机器人进行研究。传统的分拣机器人的分拣任务是一般通过示教和离线编程来完成，但是这种工作方式只适合既定位置的目标，一旦目标偏离了原来的位置，则会发生分拣任务失败的情况，导致工作效率极低。……