Road Segment Traffic Volume Estimation with ML Expansion Factors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional traffic volume estimation techniques are inaccurate and lack granularity, particularly in areas lacking traffic monitoring data, restricting their applicability and accuracy.
Innovation Solution
A system utilizing telematics data from vehicles and census data to estimate traffic volumes, applying machine learning models to determine expansion factors for road segments and larger areas, enabling accurate and granular traffic volume estimation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional traffic monitoring data from discrete locations is used, then traffic volume estimation is accurate for those specific locations, but the estimation lacks granularity and accuracy for other locations and areas
Solution Approach 1:
The patent uses telematics devices as intermediaries to collect vehicle traffic data from a distributed network of vehicles across multiple locations. These devices capture GPS coordinates, timestamps, and vehicle identifiers, serving as a mediator between physical traffic flow and digital estimation systems. This intermediary approach enables accurate traffic volume estimation for locations without traditional monitoring infrastructure by leveraging data from vehicles passing through those areas.
Solution Approach 2:
The system creates a virtual copy of traffic monitoring capabilities by utilizing telematics data from vehicles equipped with tracking devices. Instead of requiring physical monitoring infrastructure at every location, the patent replicates monitoring functionality through software-based data collection from vehicles, enabling consistent traffic estimation across diverse locations including rural and underserved areas.
2Area of stationary object
If traffic monitoring data is collected from many discrete locations, then comprehensive coverage is achieved, but the system complexity and data processing requirements increase
Solution Approach 1:
The patent implements a universal data processing framework that handles multiple data sources (telematics devices, census data, map data) and generates multiple types of outputs (traffic volume estimates, expansion factors, route analyses) through a single integrated system. The machine learning model serves as a multi-functional component that processes diverse inputs and produces standardized estimates across different geographic scales, from individual road segments to entire regions.
Solution Approach 2:
The system dynamically adjusts processing parameters based on data availability and geographic scale. Expansion factors are calculated and stored for different hierarchical levels (road segments, counties, states), allowing the system to optimize processing intensity according to the required output granularity. This parameter adaptation reduces computational complexity by pre-calculating and caching expansion factors for common query scenarios.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Disclosed herein are systems and methods for determining an expansion factor for estimating vehicle traffic. One example method comprises operating at least one processor to: receive telematics data, map data, and census data; determine, using the telematics data and the map data, an initial estimated vehicle traffic volume for each of a plurality of road segments along which a plurality of vehicles operate based on an amount of the vehicles that operate therealong; generate a total estimated vehicle traffic volume for each of the road segments along which the vehicles operate by applying a machine learning model to the initial estimated vehicle traffic volume and the census data; and determine the expansion factor for each of the road segments along which the vehicles operate based at least in part on a ratio of the initial estimated vehicle traffic volume thereof to the total estimated vehicle traffic volume.