Traffic speed data imputation method based on tensor completion

Area of Science:

• Intelligent Transportation Systems
• Data Science
• Traffic Engineering

Background:

• Missing traffic data degrades the performance of Intelligent Transportation Systems (ITS) and Advanced Traveler Information Systems (ATIS).
• Accurate traffic speed data is crucial for effective transportation management and traveler information.

Purpose of the Study:

• To develop and evaluate a novel tensor-based imputation method for estimating missing traffic speed data.
• To address the challenge of data scarcity in traffic speed datasets, which exhibit significant fluctuations.

Main Methods:

• Modeling traffic speed data using a tensor pattern.
• Employing High accurate Low Rank Tensor Completion (HaLRTC) for efficient and accurate data imputation.
• Evaluating the proposed method on the Performance Measurement System (PeMS) database.

Main Results:

• The proposed tensor-based imputation method effectively recovers missing traffic speed data, even from noisy entries.
• Experimental results demonstrate the superiority of the HaLRTC approach over existing baseline methods.
• The method accurately estimates missing data considering the inherent fluctuations in traffic speed.

Conclusions:

• The novel tensor-based imputation approach using HaLRTC is a highly effective solution for handling missing traffic speed data.
• This method enhances the reliability and performance of Intelligent Transportation Systems and Advanced Traveler Information Systems.
• The study provides a robust tool for improving traffic data quality and enabling better transportation management.