Traffic Risk Estimation via Aggregated Vehicle Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing traffic risk estimation systems struggle to accurately detect high-risk areas, leading to inappropriate activation of driving assistance functions and potential safety issues.
Innovation Solution
A traffic risk estimation device that generates aggregated data from vehicle data across a target area, allowing for accurate determination and reevaluation of traffic risks in unit and smaller areas, thereby activating driving assistance functions at appropriate times.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If driving assistance function is always activated at intersections with blind spots, then safety coverage is improved, but driver frustration increases and system reliability decreases
Solution Approach 1:
The patent applies local quality by differentiating risk assessment at different locations within the target area. Instead of uniformly activating driving assistance at all intersections with blind spots, the system evaluates aggregated dangerous phenomenon data to identify specific high-risk unit areas, and only activates assistance functions in those localized high-risk zones. This resolves the contradiction by maintaining safety coverage where needed while avoiding unnecessary activation in lower-risk areas, thereby reducing driver frustration.
Solution Approach 2:
The system changes the parameter of risk assessment from binary (blind spot presence/absence) to continuous (aggregated dangerous phenomenon frequency). By using aggregated data about dangerous phenomena occurring in unit areas and calculating traffic risk based on occurrence frequencies, the system dynamically adjusts driving assistance activation decisions. This parameter transformation enables more accurate reliability assessment while improving ease of operation through context-aware activation.
2Difficulty of detecting and measuring
If recognition sensors such as in-vehicle cameras are used, then real-time detection capability is improved, but detection accuracy of dangerous intersections remains insufficient
Solution Approach 1:
The patent merges two detection approaches: real-time detection by in-vehicle recognition sensors and historical aggregate data analysis of dangerous phenomena. The system combines information from multiple sources (sensor data + aggregated dangerous phenomenon data from multiple vehicles) to comprehensively assess traffic risk. This combination resolves the contradiction by maintaining real-time detection capability while significantly improving detection accuracy through multi-source data fusion.
Solution Approach 2:
The system performs preliminary action by pre-calculating traffic risk for each unit area based on aggregated dangerous phenomenon data before the host vehicle approaches. The server determines high-risk areas in advance and provides this information to the host vehicle, enabling the in-vehicle recognition sensor to focus verification efforts on pre-identified risky intersections. This preliminary risk assessment improves detection accuracy while maintaining real-time response capability.
3Area of stationary object
If traffic risk estimation is performed for entire target area, then comprehensive coverage is improved, but estimation accuracy for specific unit areas decreases
Solution Approach 1:
The patent applies segmentation by dividing the target area into multiple unit areas and performing separate traffic risk estimation for each unit area based on aggregated dangerous phenomenon data. The system calculates occurrence frequencies of dangerous phenomena specifically for each unit area, enabling precise local risk assessment. This segmentation resolves the contradiction by maintaining comprehensive coverage of the entire target area while achieving high estimation accuracy for each specific unit area through localized data analysis.
Data Source
AI summary
A traffic risk estimation device includes: a data generation unit that generates aggregated data regarding a predetermined dangerous phenomenon from vehicle data acquired in an entire target area including a plurality of unit areas; and a risk determination unit that determines a traffic risk for each of the plurality of unit areas based on the aggregated data.