Intersection Lane Segmentation for Turn Behavior Evaluation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques fail to separately obtain driving characteristic parameters for traveling and oncoming lanes at an intersection, which is crucial for evaluating driving behavior, especially when turning in countries with left-hand or right-hand traffic.
Innovation Solution
An information processor that acquires section position information, vehicle behavior data, and calculates vehicle position to determine whether the vehicle is on the traveling lane or oncoming lane, extracting specific driving characteristic parameters using a trained machine learning model for evaluation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If driving characteristic parameters are obtained using existing techniques, then general driving behavior can be evaluated, but separate evaluation of traveling lane and oncoming lane behavior is not possible
Solution Approach 1:
The patent segments the driving path into distinct sections (traveling lane section and oncoming lane section) using section position information. By dividing the continuous driving trajectory into discrete lane-specific segments, the system can separately acquire and evaluate driving characteristic parameters for each lane, resolving the inability to distinguish between traveling and oncoming lane behaviors in existing techniques.
2Quantity of substance
If driving characteristic parameters are acquired for all sections, then complete driving behavior data is obtained, but the complexity of processing and analyzing data increases
Solution Approach 1:
The patent extracts only the necessary driving characteristic parameters from the complete set of acquired data by using section position information to identify and select parameters specific to traveling and oncoming lanes. This extraction process filters out irrelevant data, reducing processing complexity while maintaining the completeness of lane-specific evaluation.
Solution Approach 2:
The patent applies local quality by treating different lane sections with different evaluation criteria and parameter sets. Instead of uniformly processing all driving data, the system tailors the analysis to specific local contexts (traveling lane vs. oncoming lane), optimizing processing efficiency for each section's unique characteristics.
3Measurement precision
If vehicle position is continuously tracked through the intersection, then accurate lane determination is possible, but the computational load increases
Solution Approach 1:
The patent performs preliminary action by pre-defining section position information for traveling and oncoming lanes before the vehicle actually traverses the intersection. This advance preparation of spatial reference data allows the system to quickly determine lane position during actual driving without performing complex real-time calculations, thereby reducing computational load while maintaining precision.
Data Source
AI summary
An information processor includes: a first information acquisition unit that acquires section position information for identifying the position of a first section on a traveling lane and the position of a second section on an oncoming lane; a second information acquisition unit that stores, as history information, the history of driving characteristic parameters when traveling through the intersection; a driver's vehicle position calculation unit that sequentially calculates the driver's vehicle position information when traveling through the intersection; a section determination unit that determines which of the first section and the second section the driver's vehicle position is located; and an extraction unit that extracts a specific driving characteristic parameter corresponding to at least one of the first section and the second section from the history information based on the result of the determination by the section determination unit.