GC-462 Traffic Pattern Analysis and Anomaly Detection Using Large-scale Trajectory Data

Location

https://ccse.kennesaw.edu/computing-showcase/cday-programs/fall23program.php

Streaming Media

Document Type

Event

Start Date

30-11-2023 4:00 PM

Description

With the advancement of IoT and improved computing capabilities, real-time vehicle and road user trajectories are easily accessible through advanced traffic sensing, replacing time-consuming manual checks. This study employs machine learning to analyze extensive trajectory data, focusing on anomaly detection in traffic patterns. It investigates efficient techniques for processing time-series data using an open-source dataset (InD dataset). The procedure involves data preprocessing, feature extraction, machine learning model training, and anomaly detection at 4 intersections. Irregular paths reveal abnormal driving behavior like U-turns and unexpected stops. The study highlights their impact on traffic management and safety and discusses potential applications in vehicle-to-infrastructure alert systems.

This document is currently not available here.

Share

COinS
 
Nov 30th, 4:00 PM

GC-462 Traffic Pattern Analysis and Anomaly Detection Using Large-scale Trajectory Data

https://ccse.kennesaw.edu/computing-showcase/cday-programs/fall23program.php

With the advancement of IoT and improved computing capabilities, real-time vehicle and road user trajectories are easily accessible through advanced traffic sensing, replacing time-consuming manual checks. This study employs machine learning to analyze extensive trajectory data, focusing on anomaly detection in traffic patterns. It investigates efficient techniques for processing time-series data using an open-source dataset (InD dataset). The procedure involves data preprocessing, feature extraction, machine learning model training, and anomaly detection at 4 intersections. Irregular paths reveal abnormal driving behavior like U-turns and unexpected stops. The study highlights their impact on traffic management and safety and discusses potential applications in vehicle-to-infrastructure alert systems.