Related Experiment Video
Updated: Jun 29, 2025
11:12
Driving Simulation in the Clinic: Testing Visual Exploratory Behavior in Daily Life Activities in Patients with Visual Field Defects
Published on: September 18, 2012
17.4K
Anomalous Behavior Detection in Trajectory Data of Older Drivers
Seyedeh Gol Ara Ghoreishi1, Sonia Moshfeghi1, Muhammad Tanveer Jan1
1Florida Atlantic University, Boca Raton, USA.
2023 IEEE 20Th International Conference on Smart Communities: Improving Quality of Life Using AI, Robotics and Iot (HONET)
|April 3, 2024
Summary
This study introduces an Edge-Attributed Matrix to detect abnormal driving behaviors from trajectory data. This method aids in identifying potential health issues like Mild Cognitive Impairment (MCI) and improving road safety.
Area of Science:
- Computational science
- Transportation safety
- Data science
Background:
- Anomalous behavior detection (ABD) in driving is crucial for applications like Mild Cognitive Impairment (MCI) detection and safe navigation for elderly drivers.
- Analyzing large, temporally-detailed trajectory datasets for ABD presents significant computational challenges.
Purpose of the Study:
- To develop an efficient method for identifying abnormal driving behaviors, characterized by directional deviations, hard-braking, and acceleration.
- To address the computational complexity associated with processing extensive trajectory data for ABD.
Main Methods:
- Proposing an Edge-Attributed Matrix to effectively represent key properties of temporally-detailed trajectory datasets.
- Utilizing this matrix to identify drivers exhibiting anomalous driving patterns.
Main Results:
- The proposed Edge-Attributed Matrix successfully represents critical trajectory data properties.
- Experimental results on real-world datasets confirm the approach's effectiveness in identifying abnormal driving behaviors.
Conclusions:
- The Edge-Attributed Matrix offers a computationally feasible solution for anomalous driving behavior detection.
- This approach has significant implications for public health and transportation safety applications.
