Road Closure Detection Using Probe Data and Dynamic Time Windows
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing traffic service providers face challenges in accurately detecting road closures due to incomplete or inaccurate road closure reports, which can lead to poor quality data and user experiences, and there is a need for a more reliable method to automatically detect road closures.
Innovation Solution
A system that utilizes probe data from vehicles to automatically detect road closures by constructing a closure link graph, aggregating road links into superlinks, and using a dynamic time window to calculate features for road closure probability, reducing reliance on manual efforts and enhancing coverage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If road closure reports are collected from manual sources, then coverage of road closures can be obtained, but the data quality is incomplete and inaccurate
Solution Approach 1:
The system uses probe data from vehicles as feedback to verify and correct road closure information. By continuously monitoring vehicle trajectories and detecting anomalies in traffic patterns, the system validates reported road closures and identifies false positives, thereby improving both information completeness and detection accuracy simultaneously
Solution Approach 2:
The system enables automatic self-detection of road closures through probe data analysis without relying on manual reports. The automated detection mechanism processes vehicle trajectory data to identify closure patterns, reducing dependence on external reporting sources and improving data quality through objective, data-driven detection
2Measurement precision
If automated detection using probe data is implemented, then detection accuracy is improved, but system complexity increases
Solution Approach 1:
The system leverages existing probe data infrastructure that is already collecting vehicle trajectory information for other traffic analysis purposes. By making this existing data serve multiple functions including road closure detection, the system achieves high detection accuracy without proportionally increasing system complexity
Solution Approach 2:
The system introduces a specialized analysis module that acts as an intermediary between raw probe data and road closure detection results. This modular approach processes probe data through specific algorithms to extract closure information, managing complexity by isolating the detection logic from the broader traffic data infrastructure
3Productivity
If manual detection methods are used, then system simplicity is maintained, but productivity and response speed are reduced
Solution Approach 1:
The system replaces manual detection processes with automated computational analysis of probe data. Algorithms automatically process vehicle trajectory information to detect road closures, eliminating the need for human analysts to manually review data, thereby dramatically increasing detection speed and productivity
Solution Approach 2:
The system implements continuous monitoring of probe data streams to detect road closures in real-time rather than through periodic manual checks. This continuous automated analysis ensures rapid detection and response to new closures, maintaining high productivity without requiring proportional increases in manual labor
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An approach is provided for automatic road closure detection. The approach, for example, involves designating a dynamic time window comprising one or more time epochs ending before a current time epoch. The approach also involves retrieving a first set of probe data collected from a road link during the dynamic time window. The approach further involves adjusting a size of the dynamic time window by adding or removing another time epoch ending before the current time epoch until at least one criterion related to the probe data, the dynamic time window, or a combination thereof is met. The approach further involves extracting a plurality of features from the first set of probe data, from a second set of probe data collected from the road link during the current time epoch, or a combination thereof. The approach then involves detecting a closure status of the road link based on the plurality of features.