| 基于UWB定位的邮轮乘员伴随关系发现算法 |
| |
| 引用本文： | 严思迅,吴兵,商蕾,吕洁印,汪洋.基于UWB定位的邮轮乘员伴随关系发现算法[J].交通信息与安全,2021,39(6):54-62,99. |
| |
| 作者姓名： | 严思迅 吴兵 商蕾 吕洁印 汪洋 |
| |
| 作者单位： | 1.武汉理工大学国家水运安全工程技术研究中心 武汉 430063 |
| |
| 基金项目： | 国家自然科学基金青年项目51809206工信部高技术船舶科研项目G18473CZ06深圳市科技创新委员会项目CJGJZD20200617102602006 |
| |
| 摘 要： | 准确发现邮轮内部空间乘客之间的伴随关系, 在室内环境安装UWB定位设备开展室内人员定位实验。根据UWB定位的位置数据特点, 提出结合室内位置语义的Hausdorff-DBSCAN算法以聚类邮轮乘员轨迹, 并利用LSTM神经网络对疑似伴随关系对象进行相似度变化趋势的预测。传统的Hausdorff算法在计算轨迹相似度时未考虑轨迹时序一致的问题, 引入位置语义序列能够较好地解决这个问题。改进后的Hausdorff-DBSCAN算法的输入为乘员轨迹数据集, 根据轨迹整体相似度阈值选定聚类半径, 输出具有伴随关系的乘员轨迹聚类结果; LSTM神经网络以定长时间窗口的点邻近度序列为输入, 预测后1个时刻点邻近度值, 结合轨迹相似度阈值和预测结果分析乘员伴随关系的时序变化。利用Anylogic建模单层邮轮室内环境进行乘员仿真得到的轨迹数据验证算法的有效性。改进的Hausdorff-DBSCAN算法的准确率为0.920, 召回率为0.950, F1值为0.934, 准确率高出对比算法至少5.7%, 召回率高出对比算法至少8.0%, F1值高出对比算法至少6.7%。同时LSTM在预测邮轮乘员之间相似度变化时, 收敛后的误差值能保持在3%~4%左右, 预测结果具有较高的准确性。
|
| 关 键 词： | 交通信息 邮轮乘员轨迹 伴随关系发现 DBSCAN聚类 LSTM神经网络 轨迹相似性 |
| 收稿时间： | 2021-07-31 |
A Detection Algorithm for Discovering Accompanying Relationship of Cruise Passengers Based on UWB Positioning |
| |
| YAN Sixun,WU Bing,SHANG Lei,LYU Jieyin,WANG Yang.A Detection Algorithm for Discovering Accompanying Relationship of Cruise Passengers Based on UWB Positioning[J].Journal of Transport Information and Safety,2021,39(6):54-62,99. |
| |
| Authors: | YAN Sixun WU Bing SHANG Lei LYU Jieyin WANG Yang |
| |
| Affiliation: | 1.National Engineering Research Center for Water Transport Safety, Wuhan University of Technology, Wuhan 430063, China2.School of Transportation and Logistics, Wuhan University of Technology, Wuhan 430063, China3.School of Naval Architecture, Ocean and Energy Power Engineering, Wuhan University of Technology, Wuhan 430063, China4.CIMC Intelligent Technology Co. Ltd., Shenzhen 518057, Guangdong, China |
| |
| Abstract: | UWB positioning is used in the cruise to carry out an on-board personnel location experiment to discover the accompanying relationship among passengers in the interior space of a cruise. A improved scheme based on Haussdorff-DBSCAN is proposed combined with indoor positional semantics to study the clustering of the passenger trajectories, based on the characteristics of the UWB location data. Afterward, the LSTM neural network is applied to predict the changing similarity of the suspected companions. The traditional Hausdorff algorithm does not consider the consistency of trajectory timing while calculating the trajectory similarity, and the introduction of positional semantic sequence can solve this problem. In the first phase, the passenger trajectory data set is input to the improved Hausdorff-DBSCAN algorithm, and the clustering radius is determined according to the overall similarity threshold of trajectories. The outputs are the emerging clusters of passenger trajectories in the same companion group. In the second phase, the LSTM neural network takes the point similarity sequence with a fixed time window as the input to predict the point similarity value at the next time. The sequential change of passengers accompanying relationship is analyzed by the similarity threshold and predicted results. The validity of the presented algorithm is demonstrated by the trajectory data obtained from the simulation study, which is modeled in Anylogic. The results indicate that the precision of the proposed algorithm, the recall value, and the F1 value are 0.92, 0.95, and 0.934, which are at least 5.7%, 8.0%, and 6.7% higher than the compared algorithm, respectively. The LSTM neural network also shows a promising effect in predicting the changing trend of the similarity because the loss is at a stable level from 3% to 4%. |
| |
| Keywords: | |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《交通信息与安全》浏览原始摘要信息 |
|
点击此处可从《交通信息与安全》下载全文 |
|
