Stay Point Classification Using Probe Trajectory Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing map service providers face challenges in automatically distinguishing between on-road and off-road stay points, requiring resource-intensive manual processes and human observation, which do not scale well with increasing data volumes.
Innovation Solution
A computer-implemented method that segments probe trajectories into different types using map matching confidence and merges unknown segments based on co-occurrence, applying computational geometry to identify stay points, thereby reducing the need for human classification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual processes and human observation are used to classify stay points, then classification accuracy can be maintained, but resource consumption increases and scalability deteriorates
Solution Approach 1:
The system enables automated self-classification of stay points using probe data and map matching algorithms. The computational geometry-based approach allows the system to automatically distinguish between on-road and off-road stay points without human intervention, achieving both high accuracy and scalability by eliminating the need for manual classification processes
Solution Approach 2:
The patent replaces manual human observation and classification processes with an automated computational system. By using map matching confidence values and computational geometry algorithms to process probe trajectory data, the system substitutes human labor with machine-based automated classification, maintaining accuracy while dramatically improving scalability and reducing resource consumption
2Ease of manufacture
If manual classification processes are deployed, then stay point categorization can be performed, but resource consumption and operational complexity increase
Solution Approach 1:
The system performs automated self-classification using probe data from mobile devices. By leveraging existing probe data and map matching algorithms, the system eliminates the need for external manual classification resources, reducing operational complexity and resource consumption while maintaining classification capability
Solution Approach 2:
The system uses probe data copies from mobile devices to classify stay points without requiring physical human observation. By processing data copies through computational geometry algorithms, the system achieves classification with minimal resource consumption compared to deploying personnel for manual verification
3Speed
If probe data is processed without segmentation, then processing speed may be maintained, but classification accuracy and detail deteriorate
Solution Approach 1:
The patent segments probe trajectories into distinct trajectory segments based on map matching confidence values. This segmentation allows the system to identify transitions between on-road and off-road locations, enabling precise stay point classification. The segmented approach maintains processing efficiency while significantly improving classification accuracy by analyzing meaningful portions of trajectory data
Data Source
AI summary
An approach is provided for detecting stay points from probe data. The approach, for example, involves segmenting a probe trajectory into a first trajectory segment type (e.g., on-road), a second trajectory segment type (e.g., off-road), an unknown trajectory segment type, or a combination thereof. The approach also involves merging the unknown trajectory segment into either the first trajectory segment type or the second trajectory segment type based on a co-occurrence with the unknown trajectory segment. The approach further involves segmenting the first trajectory segment type into first sub-segments and the second trajectory segment type second sub-segments based on a minimum enclosing ball or any other bounding volume with a radius less than or equal to the distance threshold. The approach further involves processing each first sub-segment and each second sub-segment to determine stay points.