Dynamic grouping of vehicle trajectories

Abstract

Vehicular traffic volume in large cities has increased in recent years, causing mobility problems; therefore,the analysis of vehicle flow data becomes a relevant research topic. Intelligent Transportation Systems monitor and control vehicular movements by collecting GPS trajectories, which provides the geographic location of vehicles in real time. Thus information is processed using clustering techniques to identify vehicular flow patterns. This work presents a methodologycapable of analyzing the vehicular flow in a given area, identifying speed ranges and keeping an interactivemap updated that facilitates the identification of possible traffic jam areas. The results obtained on threedata sets from the cities of Guayaquil-Ecuador, RomeItaly and Beijing-China are satisfactory and clearlyrepresent the speed of movement of the vehicles, automatically identifying the most representative ranges inreal time.

Network Congestion Visualization for Vehicular Communication using Streamlit Approach

“…This section provides a detailed description of the methods used to map network addresses in vehicular networks. Methods include using Streamlit, a Python library for developing interactive web applications, and using MATLAB for visualization Below we describe the complexity of each method and clarify their specific implementation, performance, and comparative aspects is visible [8] [9].…”

Method for the Identification and Classification of Zones with Vehicular Congestion

“…There are identification proposals for static clustering that use, as a basis, a fixed grid-based technique to process data streams from various features [17,18]. Although this processing method leads to the identification of different congestion patterns in a simple way [19], they have the disadvantage that they keep in the background the temporal characteristic present in the data of the trajectories; this disadvantage causes a situation where the information of the clusters can present persistent patterns that should not be present if it is analyzed temporally.…”

“…A multitude of papers have showcased diverse solutions, including a methodology for examining vehicular flow in defined zones, identifying speed ranges, and upkeeping an interactive map that stays current, aiding in the manual inspection of congestionprone regions [18]. Although this representation provides a summarized view of realtime traffic, it is essential to incorporate additional information to enrich the analysis of vehicular flow [38], such as information from the road infrastructure or information from different sensors.…”

Method for the Identification and Classification of Zones with Vehicular Congestion

“…There are identification proposals for static clustering that use, as a basis, a fixed grid-based technique to process data streams from various features [17,18]. Although this processing method leads to the identification of different congestion patterns in a simple way [19], they have the disadvantage that they keep in the background the temporal characteristic present in the data of the trajectories; this disadvantage causes a situation where the information of the clusters can present persistent patterns that should not be present if it is analyzed temporally.…”