Context-Aware Driver Scoring for Outlier Driving Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle-based systems struggle to accurately assess driver scores for vehicles with unique driving patterns and account for less safe driving behaviors, especially when multiple drivers use the same vehicle or when contextual factors influence driving decisions.
Innovation Solution
A method that utilizes telematics data from both vehicle-based systems and mobile devices to group vehicles by their driving patterns, identify outliers, and adjust driver scores based on safety metrics, considering contextual data to differentiate between safe and unsafe driving events, particularly when events are triggered by other vehicles' risky behaviors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If driver scores are calculated based on standard driving behavior metrics, then driving safety assessment is simplified and efficient, but vehicles with unique driving patterns are misclassified and scoring accuracy deteriorates
Solution Approach 1:
The patent segments the driver population into multiple clusters based on driving behavior patterns using unsupervised machine learning. Instead of applying a single standardized scoring model to all drivers, the system divides them into distinct groups (e.g., aggressive drivers, defensive drivers, average drivers) and develops tailored scoring models for each segment. This segmentation allows the system to maintain scoring efficiency while significantly improving accuracy for vehicles with unique driving patterns by evaluating each against appropriate peer benchmarks rather than a one-size-fits-all standard.
2Measurement precision
If contextual factors are considered in driver scoring, then scoring accuracy and fairness are improved, but system complexity and computational requirements increase
Solution Approach 1:
The patent performs preliminary clustering of drivers into behavior-based segments before applying contextual factor analysis. By pre-grouping drivers with similar driving patterns, the system reduces the complexity of subsequent contextual analysis, as each cluster can be evaluated with cluster-specific contextual benchmarks rather than requiring individualized contextual assessment for every driver. This preliminary segmentation simplifies the overall system architecture while maintaining the ability to incorporate contextual factors for improved accuracy.
Solution Approach 2:
The patent introduces cluster-based driving pattern profiles as intermediaries between raw driving data and final score calculations. These profiles serve as mediators that capture contextual characteristics of each driver segment, allowing the system to incorporate complex contextual factors without directly processing all raw data points for each individual driver. The intermediary profiles simplify computational requirements while preserving the ability to account for contextual variations in driving behavior.
3Measurement precision
If outlier vehicles are identified and scored separately, then scoring accuracy for unique driving patterns is improved, but the overall scoring system complexity increases
Solution Approach 1:
The patent integrates outlier detection into the clustering process itself, where unsupervised learning algorithms automatically identify and segment outlier vehicles as distinct clusters or as part of smaller sub-clusters within the overall driver population. Rather than adding a separate outlier detection layer to an existing scoring system, the segmentation approach built-in identifies unique driving patterns during the initial clustering phase, reducing overall system complexity while maintaining high accuracy for outlier vehicles.
Data Source
AI summary
A driving analysis server may be configured to receive vehicle operation data from mobile devices respectively disposed within the vehicles, and may use the data to group the vehicles into multiple groups. A driving pattern for each vehicle may be established and compared against a group driving pattern of its corresponding group to identify outliers. A driver score the outliers may be adjusted positively or negatively based on whether the outlier behaved in a manner more or less safe than its group. Further, unsafe driving events performed by the outlier that were the result of another vehicle's unsafe driving event may be ignored or positively accounted for in determining or adjusting the outlier's driver score.