Trajectory Clustering for Modal Route Determination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Mapping and navigation service providers face challenges in determining the most popular routes taken by users between origin and destination points, as calculated routes may not align with actual user behavior due to factors like road repairs or hidden issues not accounted for in traditional routing algorithms.
Innovation Solution
A system that retrieves and clusters trajectory data from probe points to identify modal routes by processing position, heading, speed, and time data from sensors, allowing for the determination of popular routes based on actual user travel patterns, which can indicate anomalies in calculated routes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If traditional routing algorithms are used to calculate routes, then routing calculation speed is improved, but accuracy of determining actual popular routes deteriorates
Solution Approach 1:
The patent uses probe data collected from multiple devices (copies of actual travel behavior) to determine modal routes, rather than relying on single calculated route predictions. This copying approach allows the system to observe actual user behavior patterns across multiple instances to identify true popular routes.
Solution Approach 2:
The system implements feedback by using actual probe data from devices to continuously update and refine route determination. The feedback loop involves collecting real travel data, analyzing it to identify modal routes, and using this information to improve future routing calculations, thereby increasing measurement precision over time.
2Measurement precision
If probe data from multiple devices is collected and processed, then accuracy of determining modal routes is improved, but system complexity increases
Solution Approach 1:
The patent segments the complex task of determining modal routes into distinct processing stages: collecting probe data from multiple devices, processing the data to identify trajectories, clustering trajectories into groups, and determining modal routes based on cluster characteristics. This segmentation reduces overall system complexity by breaking down the complex analysis into manageable steps.
Solution Approach 2:
The system merges data from multiple individual devices into a unified analysis by aggregating probe data into trajectories and then clustering these trajectories into groups. This merging approach allows the system to handle complexity through consolidation, where individual device data is combined to reveal patterns that would be invisible in single-device data.
Data Source
AI summary
An approach is provided for determining one or more modal routes between one or more origin areas and one or more destination areas based on trajectory data. The approach involves querying, by a processor, a trajectory database to retrieve one or more trajectories that contain one or more probe points located within the one or more origin areas and the one or more destination areas. The approach also involves clustering, by the processor, the one or more trajectories into one or more trajectory clusters based on a distance. The approach further involves determining, by the processor, respective counts of the one or more trajectories within each of the one or more trajectory clusters. The approach further involves presenting, by the processor, at least one of the one or more trajectory clusters as the one or more modal routes based on the respective counts.