Traffic Interpolation Using Principal Component Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional traffic information interpolation methods fail to provide high precision when the missing rate of floating car data is high, leading to inaccurate real-time traffic information, especially in areas with complex road link connections.
Innovation Solution
Performing principal component analysis on past floating car data to calculate bases, dynamically selecting bases with strong correlation to real-time data collection links, and using weighting coefficients for linear combination to estimate missing traffic data, ensuring stable and precise interpolation even with a small number of real-time data points.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional interpolation techniques using neighboring road-links are used, then interpolation is attempted, but precision deteriorates when the missing rate of floating car data is high (90% or more) because neighboring links also have missing data
Solution Approach 1:
The patent transforms the interpolation problem from spatial domain to feature space domain by performing principal component analysis. Instead of interpolating based on geographical proximity (spatial dimension), the method projects traffic data onto principal components (feature space dimension) where correlations are preserved. This allows accurate interpolation even when neighboring spatial locations have missing data, as the feature space representation captures the underlying traffic patterns across the entire network.
Solution Approach 2:
The patent changes the parameters used for interpolation from simple spatial coordinates to principal component scores. By transforming the original traffic data into principal component space, the method identifies the most significant variance patterns in traffic behavior. Interpolation is then performed in this transformed parameter space where the relationships between different road-links are captured more effectively, leading to higher precision even with high missing data rates.
2Area of stationary object
If interpolation is performed using remote road-links when spatial missing rate is high, then coverage is improved, but precision deteriorates in areas with complicated road link connections
Solution Approach 1:
The patent resolves this contradiction by moving the interpolation operation from the spatial dimension to the feature space dimension. In feature space, all road-links are represented by their principal component scores, allowing the method to utilize data from any road-link regardless of its spatial distance. The principal component analysis captures the correlated variations across the entire network, enabling accurate interpolation anywhere in the coverage area without being constrained by spatial proximity or road network complexity.
3Loss of information
If statistical processing of past FCD is used, then interpolation with high missing rate is possible, but the processed traffic information does not exactly reflect real-time traffic information
Solution Approach 1:
The patent applies preliminary action by pre-calculating the principal components from historical traffic data, but performs the actual interpolation using real-time data when available. The principal component analysis is performed offline on past FCD to establish the basis for interpolation, but the weighting coefficients are calculated using real-time FCD projection. This hybrid approach enables handling of high missing rates while maintaining real-time accuracy, as the method adapts to current traffic conditions rather than relying solely on historical patterns.
Solution Approach 2:
The patent incorporates feedback by using real-time floating car data to calculate weighting coefficients for the principal components. Even when real-time FCD is sparse, the system projects available real-time data onto the pre-established principal components and uses these projections to weight the interpolation. This feedback mechanism ensures that the interpolation reflects current traffic conditions rather than merely reproducing historical patterns, maintaining reliability while handling high missing rates.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
In a traffic information system, the principal component analysis of the floating car data collected in the past is performed for each traffic area. From among the bases representing the traffic data collected on the road-links in the traffic area, the bases which have strong correlation to the road-links on which real-time traffic data were collected are selected. The weighting coefficients for the selected bases are calculated by projecting the real-time traffic data onto the selected bases. The traffic estimation data are calculated by linearly combining the selected bases with the obtained weighting coefficients as the coefficients of the respective bases. The calculated traffic estimation data are used for the interpolation of the road-links on which the real-time traffic data were not collected.