Vehicle Path Travel Time Distribution via Copula-Based Segment Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for calculating travel time distribution for driver assistance systems are inaccurate and time-consuming, making them impractical for real-world applications.
Innovation Solution
A computer-implemented method using dependent discrete convolution (DDC) with copula-based correlation to estimate travel time distribution by incorporating the dependency between road segments, utilizing historical data from a vehicle fleet to calculate the path travel time distribution efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional statistical modeling methods are used to estimate path travel time distribution, then measurement precision may be improved, but calculation time increases significantly making it impractical for real-world applications
Solution Approach 1:
The patent segments the path into multiple road segments and calculates travel time distribution for each segment independently using historical data from probe vehicles. By dividing the overall path calculation into smaller segment-level calculations, the system achieves accurate path travel time distribution estimation while reducing computational complexity and calculation time compared to traditional holistic statistical modeling approaches.
2Measurement precision
If dependent discrete convolution with copula-based correlation is used to account for segment dependencies, then travel time distribution accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent introduces copula functions as an intermediary mathematical tool to model the dependency relationships between adjacent road segments. The copula-based correlation framework allows the system to account for spatial dependencies and traffic flow correlations between segments without requiring complex multi-dimensional integrations. This intermediary approach enables accurate path travel time distribution calculation while keeping computational complexity manageable through efficient convolution algorithms.
3Measurement precision
If comprehensive historical data from multiple vehicles is collected and processed, then travel time distribution accuracy is improved, but data processing time and system complexity increase
Solution Approach 1:
The patent performs preliminary processing of historical probe vehicle data to pre-calculate and store travel time distributions for individual road segments. By preparing segment-level travel time distributions in advance and storing them for quick retrieval, the system avoids re-processing raw historical data during real-time path calculation. This preliminary action significantly improves data processing efficiency while maintaining accurate path travel time distribution estimation through the pre-processed segment data.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
The invention relates to a computer implemented method for calculating a travel time distribution (TTD) of a vehicle path (1) for a driver assistance function of the vehicle (3) comprising calculating (S4) a dependence-factor of a dependency between the first road segment (α) and the second road segment (β), wherein the dependence-factor incorporates a copula-based correlation between the first road segment (α) and the second road segment (β), and performing (S5) a discrete convolution dependent upon the dependence-factor to calculate the travel time distribution (TTD) of the vehicle path (1). The invention further relates to a system for calculating a travel time distribution (TTD) of a vehicle path (1) for a driver assistance function of the vehicle. The invention further relates to a computer program and a computer-readable data carrier.