摘要
为了解决机场行李在转运过程中存在的装箱问题,从码放策略、学习算法模型两个方面进行研究,提出基于深度稀疏最小二乘支持向量机模型(Deep Sparse Least Squares Support Vector Machine, DSLSSVM)的行李码放策略。该策略包括评价和决策两个部分。在评价时,通过将行李车与行李离散化,建立两者的数学关系,从而得到评价行李码放位置优劣的评估参数;在决策时,利用深度稀疏最小二乘支持向量机模型学习工人的码放经验,将决策问题转化为二分类问题,选择最佳码放位置进行码放。仿真实验表明:该算法可以达到较高的分类精度,并且能够得到比较理想的装箱效果,具有较强的实用性,满足机场行李高效运输的要求。
In order to solve the problem of packing in the process of airport baggage transfer, this paper studies the stacking strategy and learning algorithm model, and proposes a strategy of baggage stacking based on the deep sparse least squares support vector machine(DSLSSVM). This strategy included evaluation and decision-making. In the evaluation, by discretizing the luggage car and the baggage, we established the mathematical relationship between them, so as to obtain the evaluation parameters of the advantages and disadvantages of luggage stacking position. In the decision-making process, we used the deep sparse least squares support vector machine model to learn the experience of workers. The decision-making problem was transformed into binary classification problem, and the best position was selected for stacking. The simulation results show that the algorithm can achieve higher classification accuracy, and get ideal packing effect. It has strong practicability, and meets the requirements of efficient airport baggage transportation.
作者
洪振宇
张聪
Hong Zhenyu;Zhang Cong(School of Aeronautical Engineering,Tianjin 300300,China)
出处
《计算机应用与软件》
北大核心
2022年第7期44-51,66,共9页
Computer Applications and Software
基金
科技部重点研发计划项目(2018YFB1601200)
中央高校基本科研业务费项目中国民航大学专项(3122018D038)。
关键词
支持向量机
码放策略
最小二乘
稀疏化
Support vector machine
Stacking strategy
Least square
Sparsity
