Supervised Point Map Matching for GPS Heading and Distance Errors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Navigation systems face inaccuracies in locating vehicles due to errors in GPS data, particularly in challenging environments like urban areas and mountainous regions, leading to incorrect map matches which can result in severe consequences.
Innovation Solution
A method using kernel density estimation to generate probability weights for point-based map matching algorithms, combining historical probe data to improve the accuracy of map matching by distinguishing between accurately and erroneously matched positional points, and utilizing these weights in real-time map matching processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS receivers are used to obtain positional data, then location information can be acquired, but measurement precision deteriorates due to satellite geometry, multipath, and atmospheric effects
Solution Approach 1:
The patent introduces map matching as an intermediary process between raw GPS data and final location determination. The system uses probe reports containing positional data and heading data, processes them through map matching algorithms to identify the most likely road segment, and outputs a corrected location. This intermediary step filters out GPS errors by constraining possible locations to valid road positions.
Solution Approach 2:
The system implements feedback by using historical probe reports from multiple vehicles to continuously improve map matching accuracy. The accumulated probe data is used to refine the identification of accurately versus erroneously matched points, and this refined information feeds back into improving future map matching operations. The system also uses the identified accurate matches to correct GPS inaccuracies in real-time operations.
2Measurement precision
If path based map matching algorithms are used to improve accuracy, then measurement precision improves, but device complexity and processing time increase
Solution Approach 1:
The patent performs preliminary actions by using path-based map matching on historical probe reports during off-line processing to establish ground truth data. This preliminary analysis identifies accurately versus erroneously matched points, which are then used to generate probability density distributions and kernel density estimates. These pre-computed statistical models are stored and used during real-time operations, avoiding the need to run complex path-based algorithms during real-time processing.
Solution Approach 2:
The system creates simplified copies of the complex path-based map matching results in the form of probability density distributions and kernel density estimates. Instead of storing and processing the full path-based algorithm outputs during real-time operations, the system uses these statistical summaries as compact representations that capture the essential matching characteristics. This copying approach maintains accuracy while reducing real-time computational complexity.
3Measurement precision
If more satellites are available for GPS reception, then measurement precision improves, but the system becomes more vulnerable to multipath and atmospheric effects
Solution Approach 1:
The patent converts the harmful effect of GPS inaccuracies (caused by multipath and atmospheric interference) into a beneficial training opportunity. By collecting probe reports that contain these errors and processing them through map matching, the system identifies patterns in erroneous versus accurate matches. These patterns are used to create probability density distributions that characterize error behavior. The system then uses these distributions to correct future GPS readings, effectively turning past errors into a corrective mechanism.
Data Source
AI summary
System and methods are provided for a supervised point map matcher. The supervised point map matcher learns parameters from historical data that provide insight into the optimal probabilistic metrics that inform the bias of probes heading and distance for segments on the roadway. Probability weights for segments are generated. A more accurate path based map matching algorithm is used to identify direction and heading errors in the historical probe data. Values for the probability weights are calculated using kernel density estimation and a gaussian probability density function. The probability weights are used to improve the real time performance of the point map matcher. A confidence value is calculated as a function of the probability weights and provided with the map matched results.