Estimating False Positive Road Event Reports
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Navigation and mapping service providers face challenges in differentiating between true and false road event reports, particularly slippery road reports, from a large number of vehicles in real-time, leading to inaccurate data and potential safety issues.
Innovation Solution
A system that estimates false positive reports by comparing data from two groups of vehicles, one considered non-problematic and another problematic, using statistical processing to calculate the percentage of defective vehicles generating false reports, and adjusts road event data accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If road event reports are collected from a large fleet of vehicles in real-time, then the quantity of data increases, but the accuracy of the data decreases due to false positive reports from defective vehicles
Solution Approach 1:
The patent creates a virtual model (copy) of the fleet's reporting behavior by selecting a representative subset of vehicles. This subset model is used to estimate the false positive rate for the entire fleet, allowing accurate data filtering without processing every individual report from every vehicle. The subset accurately replicates the statistical characteristics of the full fleet's detection patterns.
Solution Approach 2:
The system uses the road event reports from the fleet vehicles themselves to identify and filter false positives. By analyzing the patterns in the reports generated by the fleet, the system automatically determines which vehicles are producing false positives and adjusts the overall data accuracy without requiring external validation or manual inspection of each report.
2Measurement precision
If all vehicle sensor reports are processed to identify false positives, then the accuracy of differentiation between true and false reports improves, but the computational complexity and time required increases significantly
Solution Approach 1:
Instead of processing reports from all vehicles in the fleet, the system creates a representative subset model that captures the essential statistical patterns of false positive generation. This subset model requires significantly less computational resources while maintaining the ability to accurately estimate the false positive rate for the entire fleet.
Solution Approach 2:
The patent applies partial action by processing only a representative subset of vehicle reports rather than all reports from all vehicles. This subset is sufficient to statistically determine the false positive rate, achieving the necessary accuracy without the excessive computational burden of analyzing every single report from the entire fleet in real-time.
3Productivity
If a representative subset of vehicles is selected to estimate false positive rates, then the processing time and computational resources are reduced, but the reliability of the estimation may decrease
Solution Approach 1:
The system carefully constructs a subset model that accurately copies the statistical characteristics of the full fleet. By selecting vehicles and routes that represent the diversity of the entire fleet's operating conditions, the subset reliably replicates the false positive generation patterns, ensuring that estimates derived from the subset are trustworthy when applied to the full fleet.
Solution Approach 2:
The patent adjusts parameters such as the size of the subset, the selection criteria for vehicles, and the time period analyzed to optimize both processing speed and estimation reliability. By carefully controlling these parameters, the system achieves fast processing through subset analysis while maintaining sufficient reliability for practical decision-making about fleet data quality.
Data Source
AI summary
An approach is provided for estimating false positive reports of detectable road events. For example, the approach involves determining a first number of road reports from a fleet of vehicles, and operating in a geographic area during a first time period and a second number of road reports from the fleet operating in the geographic area during a second time period. The first number of road reports and the second number of reports relate to a road event detected by vehicle sensors. The approach further involves computing a difference between the first number and the second number. The approach further involves determining a percentage of defective vehicles in the fleet based on the difference. The defective vehicles are defective with respect to a detection of the road event. The approach further involves providing the percentage of defective vehicles as an output.