Split Lane Traffic Jam Detection via Probe Data Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems struggle to accurately identify and remediate split lane traffic jams due to the inability to determine the specific lane traveled by probe vehicles, as GPS location errors exceed lane widths, making it difficult to pinpoint intersections causing these jams.
Innovation Solution
The method involves receiving and clustering pre-intersection and post-intersection probe data from multiple vehicles to determine traffic level indicators, identifying differences that satisfy a threshold to determine if an intersection is causing a split lane traffic jam, and providing notifications for route recalculations or traffic management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS location data is used to determine probe vehicle position, then location can be obtained, but the precision is insufficient to identify specific lanes
Solution Approach 1:
The patent introduces an intermediary clustering system that bridges GPS location data and lane identification. Instead of directly using imprecise GPS coordinates to determine lanes, the system clusters probe data from multiple vehicles in the same geographic area to infer lane-level traffic patterns. This intermediary clustering process recovers the lost lane travel information that cannot be obtained from individual GPS points alone.
2Measurement precision
If probe data from multiple vehicles is collected to improve accuracy, then more data is available, but data processing complexity increases
Solution Approach 1:
The patent segments the probe data processing into distinct clusters based on geographic location and traffic patterns. By dividing the data into manageable clusters representing different road segments and intersections, the system can process large volumes of probe data from multiple vehicles without overwhelming computational complexity. Each cluster is analyzed independently to detect traffic jams, making the overall processing task tractable.
3Loss of time
If real-time traffic monitoring is implemented, then traffic jams can be detected promptly, but system resource consumption increases
Solution Approach 1:
The patent implements periodic action by monitoring traffic conditions at specific time intervals and updating traffic jam detections based on changing patterns. Rather than continuously processing all probe data in real-time, the system periodically analyzes clustered data to identify emerging traffic jams. This approach reduces system resource consumption while still providing timely detection of traffic conditions changes.
Data Source
AI summary
A plurality of instances of pre-intersection and post-intersection probe data are received. Each instance of pre-intersection probe data corresponds to traveling along a pre-intersection road segment before traveling through an intersection. Each instance of post-intersection probe data corresponds to traveling along a post-intersection road segment following traveling through the intersection. Instances of pre-intersection probe data are clustered into pre-intersection clusters based on a post-intersection road segment identified by the corresponding instance of post-intersection probe data. Instances of post-intersection probe data are clustered into post-intersection clusters based on the post-intersection road segment identified thereby. A traffic level indicator is determined for each cluster. A traffic level indicator difference is determined for each pair of corresponding pre-intersection and post-intersection clusters. Responsive to determining that at least one traffic level indicator difference is greater than a threshold traffic level indicator difference, the intersection is identified as experiencing a traffic jam.