物联网技术在立体库货物盘点定位方面的应用研究

发布时间：2018-06-30 18:59

本文选题：物联网技术 + 立体库　；参考：《天津科技大学》2016年硕士论文

【摘要】：立体库是现代物流高速发展的产物,它的使用大大提高了物流企业的运作效率和现代化水平,提升了仓库空间的利用率,使得在同样空间内可存储货物种类、数量极大提高。但是,在仓储量提高的同时对于在库货物的管理也有待加强。盘点是在库货物管理中的一项重要内容,对于货物的实时、准确、高效盘点是企业确定仓储策略的基础,传统意义上的盘点一般只是对库存货物的实际数量进行清查、清点,以确保账面与实际相符合,然而在实际应用中,除了货物数量外,货物存放的位置也直接影响着盘点结果,从而影响在库货物信息和仓储运作等,因此,本文提出盘点要在数量核查的基础上,增加对于位置信息的检查,即盘点定位。对于立体库而言,总体积庞大,货架高,货格多,库存货物种类纷繁,因此对于立体库所存货物的盘点更加困难,传统人至货前盘点的方式效率低,实时性差,易错易漏,利用堆垛机移动人员位置安全性差,更重要的是即便将盘点设备放置堆垛机上扫码读数,不用人至货前,也只能保证盘点数量,而无法做到检查货物所放位置是否正确,为了解决此问题,本文拟在立体库中引用室内定位技术,建立自动盘点系统。首先通过查阅文献资料,对比分析室内定位技术,确定采用物联网技术中非视距、低成本、高精度、远距离的RFID定位技术,以电子标签和阅读器作为通信载体。定位需要知道货物与读写器之间的确切距离和位置坐标,这要通过定位算法来计算,因此本文又从测距和位置计算两部分对基于RFID的定位算法进行深入研究,测距部分采用RSSI测距算法,利用读写器读到的标签的信号强度,通过对数距离损耗模型算出确切距离d,位置计算部分,本文采用三边测量法,将其拓展到三维空间中,利用测距所得距离进一步计算出位置坐标,建立该基本定位算法的流程,完成定位。然后本文分析误差来源,环境参数的估计和已知节点的随机选取使定位算法误差加大,需要对基本算法进行优化,因此,提出"邻近节点法"和"已知节点通信法"对基本定位算法进行双重优化,并再一次建立优化后定位算法流程。最后,采用MATLAB软件对盘点定位系统进行仿真实验,编程计算定位误差,结果表明算法优化后,定位精度有明显提高,优化效果明显,但是由于室内障碍物多,且节点密度较低等,优化结果还不能完全满足立体库货物的盘点定位精度要求。分析了未完全达到目标的原因,有待加强研究。未来,可将本文建立的定位系统用于动态网络中,更加满足实际应用,同时还要在定位肓区深入研究,做到最大覆盖,真正做到将定位用于实处。
[Abstract]:Three-dimensional warehouse is the product of the rapid development of modern logistics. Its use has greatly improved the operational efficiency and modernization level of logistics enterprises, improved the utilization rate of warehouse space, made the goods can be stored in the same space, the number of goods can be greatly increased. However, at the same time as the increase of storage volume, the management of goods in storage also needs to be strengthened. Inventory is an important content in warehouse cargo management. The real-time, accurate and efficient inventory of goods is the basis for enterprises to determine the storage strategy. In traditional sense, inventory is usually just an inventory of the actual quantity of goods in stock. In order to ensure that the books are in line with the actual situation, however, in practical applications, in addition to the quantity of the goods, the location of the goods stored also directly affects the inventory results, thus affecting the inventory information and the operation of the warehousing, etc., therefore, In this paper, it is suggested that the check of position information should be added on the basis of quantity check, that is, inventory location. For the three-dimensional storehouse, the total volume is huge, the shelf is high, the goods are many and the kinds of goods in stock are numerous. Therefore, it is more difficult for the inventory of the goods stored in the three-dimensional storehouse. The traditional way of checking the goods before the person arrives at the goods is inefficient, the real time is poor, the mistake is easy to leak, The use of stacker to move people's position is poor security, more importantly, even if the inventory equipment is placed on the stacker to scan the code reading, without the need to arrive at the cargo, but also can only ensure the number of counts, but not to check the location of the goods is correct, In order to solve this problem, this paper introduces the indoor positioning technology in the stereo database and establishes an automatic inventory system. Firstly, by consulting the literature, comparing and analyzing the indoor positioning technology, we decide to adopt the technology of "Internet of things", which is non-line-of-sight, low-cost, high-precision, long-distance RFID positioning technology, with electronic tags and readers as the communication carrier. Location needs to know the exact distance and position coordinate between the cargo and the reader, which must be calculated by the location algorithm. Therefore, the location algorithm based on RFID is studied in this paper from two parts: ranging and location calculation. RSSI ranging algorithm is used in the ranging part, and the signal intensity of the tag read by the reader is used. The exact distance d and position are calculated by the distance loss model. In this paper, the trilateral measurement method is used to extend it to the three-dimensional space. The location coordinates are further calculated by using the distance obtained from the distance measurement, and the flow of the basic location algorithm is established to complete the location. Then this paper analyzes the error sources, the estimation of environmental parameters and the random selection of known nodes to increase the error of the localization algorithm, which requires the optimization of the basic algorithm. In this paper, the "neighborhood node method" and "known node communication method" are proposed to double optimize the basic localization algorithm, and the algorithm flow after optimization is established again. Finally, MATLAB software is used to simulate the inventory positioning system, and the positioning error is calculated by programming. The results show that the positioning accuracy is obviously improved and the optimization effect is obvious after the optimization of the algorithm, but there are many obstacles in the room. And the node density is low, so the optimization results can not meet the inventory positioning accuracy requirements. The reasons why the target has not been completely achieved are analyzed, and the research needs to be strengthened. In the future, the positioning system can be used in the dynamic network, which is more suitable for practical application. At the same time, it should be deeply studied in the location terminally area to achieve the maximum coverage and the real location can be used in practice.
【学位授予单位】：天津科技大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：F253;TP391.44;TN929.5

【相似文献】