Scholarlink

<p>Each country can develop its own traffic information notification system by using extremely rich information data from journey data mounted on vehicles participating in traffic for sustainable development purposes instead of just using Google applications (Google Map) because of the great benefits it brings, such as data control, improved reliability and security, and domestic technology development. This study aims to provide a comprehensive model to establish a road user map based on large volumes of data generated by the Global Navigation Satellite System (GNSS) tracking devices mounted on means of transport. Specifically, this study provides a method for processing and analyzing big data sources, which are GNSS-tracking data of registered vehicles for the transport business. Demonstrating the enhanced- Long Short-Term Memory (enhanced -LSTM) as GA-LSTM (Genetic Algorithm- Long Short-Term Memory) method, that the original LSTM network was optimized with a genetic algorithm to improve performance, is a suitable solution for forecasting the velocity of urban roads after comparing this method with other popular Traffic Speed Forecasting (TSF) methods. Furthermore, this research outlines the initial fundamental steps for establishing a traffic forecasting map for an arterial road, which can serve as a scientific foundation for researchers to develop traffic forecasting maps for entire urban regions in future studies according to the abundant data sources from GNSS tracking data.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Development of a Sustainable National Traffic Information Notification System: A GNSS-Based with Enhanced-LSTM for Urban Road Traffic Speed Forecasting

  • 期刊论文

  • 发表日期：

    2025年2月12日

查看全文
  • Do Van Manh,
  • Dinh Tuan Hai

摘要

Each country can develop its own traffic information notification system by using extremely rich information data from journey data mounted on vehicles participating in traffic for sustainable development purposes instead of just using Google applications (Google Map) because of the great benefits it brings, such as data control, improved reliability and security, and domestic technology development. This study aims to provide a comprehensive model to establish a road user map based on large volumes of data generated by the Global Navigation Satellite System (GNSS) tracking devices mounted on means of transport. Specifically, this study provides a method for processing and analyzing big data sources, which are GNSS-tracking data of registered vehicles for the transport business. Demonstrating the enhanced- Long Short-Term Memory (enhanced -LSTM) as GA-LSTM (Genetic Algorithm- Long Short-Term Memory) method, that the original LSTM network was optimized with a genetic algorithm to improve performance, is a suitable solution for forecasting the velocity of urban roads after comparing this method with other popular Traffic Speed Forecasting (TSF) methods. Furthermore, this research outlines the initial fundamental steps for establishing a traffic forecasting map for an arterial road, which can serve as a scientific foundation for researchers to develop traffic forecasting maps for entire urban regions in future studies according to the abundant data sources from GNSS tracking data.