...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>International Journal of Performability Engineering >Pattern Knowledge Discovery of Ship Collision Avoidance based on AIS Data Analysis
【24h】

Pattern Knowledge Discovery of Ship Collision Avoidance based on AIS Data Analysis

机译：基于AIS数据分析的船舶碰撞避免模式知识发现

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Maritime traffic pattern is very important for intelligent ship collision avoidance applications, as it can help provide decision support to avoid ship collision accidents and reduce casualties. There has been a large amount of Automatic Identification System (AIS) data saved by ports or management departments. If these data can be processed and analyzed scientifically to provide an early warning with appropriate collision avoidance measures, injuries or more serious results from maritime traffic may be reduced or eliminated. Our focus is to synthesize ship behaviors of interest in a clear and effective way based on automatic preprocessing and analyzing original static AIS data. One improved DBSCAN algorithm is first called to reduce the data scale and discover important data points. Then, from the perspective of Own ship, seven patterns including course change and speed change are defined to be discovered. For each special pattern, the space collision risk DCPA (distance to closest point of approach) and time collision risk TCPA (time to closest point of approach) at the beginning time and ending time are computed to confirm its situation as heading on, crossing, or overtaking with other ships in sight of one another. This unsupervised learning approach will help discover traffic pattern knowledge in current trajectories and provide decision support for future route design or anomaly analysis.
机译：海上流量模式对于智能船舶碰撞避免应用非常重要，因为它可以帮助提供决策支持，以避免船舶碰撞事故并减少伤亡。港口或管理部门保存了大量的自动识别系统（AIS）数据。如果可以科学地处理和分析这些数据，以便提供适当的碰撞避免措施的预警，可能会减少或消除来自海运流量的伤害或更严重的结果。我们的重点是基于自动预处理和分析原始静态AIS数据，以明确而有效的方式综合兴趣的船舶行为。首先调用一种改进的DBSCAN算法以减少数据刻度并发现重要数据点。然后，从自己的船舶的角度来看，定义了七种模式，包括课程变化和速度变化。对于每个特殊模式，空间碰撞风险DCPA（到最近的接近点）和时间碰撞风险TCPA在开始时间和结束时间的时间碰撞风险TCPA（最近的接近点），以确认其现状，交叉，交叉，或者与其他船只一起超越彼此。这种无监督的学习方法将有助于发现当前轨迹中的交通模式知识，并为未来的路线设计或异常分析提供决策支持。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号