GPS Map-Matching Using Shifting Fields and Space Units
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current GPS map-matching methods face challenges in accurately identifying the trajectory of moving objects on road maps due to noisy GPS data and failure to consider driver preferences and road attributes, leading to incorrect path selection.
Innovation Solution
The method constructs GPS shifting fields for road segments based on historical data and driver preferences, partitions the road map network into space units, and calculates probabilities for each road segment to determine the most likely trajectory, using a virtual network structure and spatial indexing to improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional GPS map-matching methods are used, then the process is simple and fast, but the accuracy is low due to noisy GPS data and failure to consider driver preferences and road attributes
Solution Approach 1:
The system performs preliminary actions by constructing GPS shifting fields for road segments based on historical data before actual map-matching occurs. These shifting fields pre-characterize the statistical distribution of GPS errors for each road segment, allowing the system to account for noisy GPS data and driver preferences in advance, thereby improving trajectory identification accuracy without adding excessive complexity during real-time matching
Solution Approach 2:
The map-matching process is segmented into distinct functional components: GPS shifting field constructor, space unit partitioner, virtual network constructor, and most possible path searcher. Each component handles a specific aspect of the problem (noise characterization, spatial indexing, probability calculation, path determination), which improves overall system accuracy while managing complexity through modular design
2Measurement precision
If GPS shifting fields are constructed based on historical data and driver preferences, then the accuracy of trajectory identification is improved, but the data processing time and computational resources increase
Solution Approach 1:
GPS shifting fields are constructed in advance using historical GPS data and driver preference information, storing the statistical characteristics of GPS errors for each road segment. This preliminary processing allows the system to quickly query pre-computed shifting field data during actual map-matching operations, improving road segment identification accuracy while minimizing real-time processing time
Solution Approach 2:
The road map network is partitioned into space units, and GPS shifting fields are constructed and stored for each space unit separately. This segmentation allows the system to process and store data in manageable chunks, reducing the computational burden during real-time operations while maintaining high accuracy through localized, detailed shifting field characteristics
3Productivity
If the road map network is partitioned into space units with virtual network construction, then the efficiency of path search is improved, but the system complexity increases
Solution Approach 1:
The road map network is segmented into discrete space units, with virtual networks constructed for each unit. This segmentation enables efficient spatial indexing and allows the most possible path searcher to quickly locate and process only relevant road segments containing GPS readings, dramatically improving path search efficiency while managing complexity through structured, modular organization
Solution Approach 2:
Virtual networks serve as intermediary structures between the physical road map network and the GPS matching process. These virtual networks contain probability information and candidate road segments for each space unit, acting as a mediator that simplifies the complex task of searching through the entire road network by pre-organizing data according to spatial units and GPS shifting field characteristics
4Ease of operation
If candidate road segments are selected based on GPS proximity without considering GPS shifting patterns, then the selection process is simple, but the probability of selecting incorrect road segments increases
Solution Approach 1:
The system changes the selection criteria from simple GPS proximity to a probability-based approach that incorporates GPS shifting field characteristics. Instead of selecting candidates solely based on geometric distance, the system uses the shifting fields to weight and rank candidate road segments according to the likelihood that GPS errors would produce the observed readings, thereby improving selection reliability while maintaining reasonable operational simplicity
Data Source
AI summary
An embodiment of the invention provides a method that identifies GPS shifting fields for road segments, the GPS shifting fields including areas around the road segments that include false GPS readings of objects that traveled on the road segments. The GPS shifting fields can be revised with a road segment attribute and/or at a driver preference. The road map network can be partitioned into space units, where the road map network can include at least two GPS readings of the object. For each space unit that includes a road segment, a probability that the object was located on the road segment can be calculated for each road segment based on the GPS readings of the object and the GPS shifting fields. The trajectory of the object can be determined based on the computing of the probabilities.