Lane-Level Dangerous Road Strand Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current navigation systems and location-based services lack the ability to accurately identify and quantify dangerous conditions on individual road lanes, which can lead to increased driving errors and accidents, especially in developing markets with poor road conditions.
Innovation Solution
A method and apparatus that utilize probe data analytics to calculate changes in velocity and cluster these values to identify dangerous road strands, which are then stored in a geographic database for navigation systems to provide high-precision lane-level warnings and route adjustments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If navigation systems use general road-level data without lane-level analysis, then the system complexity is low, but the measurement precision of dangerous conditions is insufficient
Solution Approach 1:
The patent divides the roadway into multiple lanes and further segments each lane into strands, allowing dangerous conditions to be identified at the lane-strand level rather than treating the entire road as a single unit. This segmentation enables precise identification of which specific lane segments have dangerous conditions while keeping other lanes unaffected by the analysis complexity
Solution Approach 2:
The patent extracts only the necessary probe data elements (velocity, position, timestamp) required for dangerous condition detection from the overall vehicle telemetry data. By selecting and extracting only the relevant subset of data needed for lane-level danger analysis, the system achieves high measurement precision without processing unnecessary information that would increase complexity
2Reliability
If the system analyzes probe data for all lanes, then the coverage of dangerous condition detection is complete, but the loss of time for data processing increases
Solution Approach 1:
The patent segments the multi-lane roadway into individual lane strands and processes probe data for each lane separately. This allows the system to maintain complete coverage across all lanes while processing smaller, manageable subsets of data for each lane independently, reducing the overall processing time compared to analyzing all lanes as a single large dataset
Solution Approach 2:
The patent applies clustering algorithms to identify dangerous lane strands by focusing on patterns of sudden velocity changes in probe data. Rather than analyzing every single data point in detail, the system uses clustering to identify representative dangerous patterns, achieving reliable detection coverage while reducing processing time through efficient pattern recognition
3Measurement precision
If the system stores detailed lane-level dangerous condition data, then the measurement precision for navigation warnings is high, but the quantity of data stored increases
Solution Approach 1:
The patent stores dangerous condition data at the lane-strand level rather than for entire roads or large geographic areas. This segmentation allows the system to maintain high measurement precision for navigation warnings by storing detailed information only for the specific lane strands where dangerous conditions exist, while avoiding storage of redundant data for safe lanes
Solution Approach 2:
The patent applies different data storage levels to different road segments based on their dangerous condition characteristics. Dangerous lane strands receive detailed lane-level danger data storage, while normal lanes use standard road-level data. This local quality approach maintains high precision where needed while minimizing overall data storage volume
Data Source
AI summary
An apparatus and method are disclosed for calculation of a dangerous road strand of a roadway. Probe data for a roadway having multiple lanes is identified. A subset of the probe data for a predetermined lane is selected and values for a change in velocity for multiple sequences in the subset of probe data are calculated. Based on clustering for the change in velocity and a danger value, the dangerous road strand is identified. The dangerous road strand is stored in a geographic database in association with the lane for the roadway.