Driver Assist Apparatus Risk Estimation for Cutting-In Vehicles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing driver assist systems are ineffective in providing smooth alerts for vehicles cutting into the expected course of a subject vehicle, as they often generate unnecessary alarms when including adjacent lane vehicles as targets, leading to frequent and sudden alerts.
Innovation Solution
A driver assist apparatus that calculates a risk estimation value based on the lateral position of adjacent vehicles, assuming they will move into the subject vehicle's expected course, to select appropriate assist content and determine if the vehicle cuts into the expected course, enabling smooth driver assistance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the driver assist system includes adjacent lane vehicles as alert targets, then the system can detect potential cutting-in risks, but it generates unnecessary alarms and frequent sudden alerts
Solution Approach 1:
The system changes the parameter of risk assessment by introducing a virtual trajectory model. Instead of simply detecting adjacent vehicles, it calculates whether these vehicles will actually cut into the subject vehicle's lane by comparing their trajectories with the subject vehicle's expected path. This parameter change from simple proximity detection to trajectory-based prediction resolves the contradiction by filtering out vehicles that won't actually cause conflict.
Solution Approach 2:
The system performs preliminary action by pre-calculating the expected course of the subject vehicle and comparing it with potential cutting-in trajectories before actual cutting occurs. This advance prediction allows the system to distinguish between vehicles that will and won't cut in, preventing unnecessary alarms while maintaining reliable detection of actual risks.
2Reliability
If the system alerts for all adjacent lane vehicles, then potential risks are covered, but the alert frequency becomes too high and sudden
Solution Approach 1:
The system changes the detection parameter from simple spatial proximity to temporal-spatial trajectory analysis. By calculating whether adjacent vehicles' paths intersect with the subject vehicle's expected course, it maintains complete risk detection for actual cutting-in scenarios while filtering out vehicles that maintain safe trajectories, thus reducing alert frequency and improving smoothness.
Solution Approach 2:
The virtual trajectory model acts as an intermediary between raw vehicle detection data and alert generation. This intermediary layer processes the detected adjacent vehicles through trajectory calculation and comparison, selectively passing only those with actual cutting-in risk to the alert system, thereby reducing unnecessary alerts while maintaining detection completeness.
3Loss of time
If the system calculates risk for vehicles not on the expected course, then cutting-in risks are detected earlier, but calculation complexity increases
Solution Approach 1:
The system performs preliminary calculation by pre-defining the subject vehicle's expected course and establishing trajectory comparison criteria before actual cutting-in occurs. This allows early detection of potential risks without requiring complex real-time calculations during the cutting maneuver itself, as the framework for assessment is already in place.
Solution Approach 2:
The system changes the calculation parameter from monitoring only vehicles on the expected course to calculating trajectories of adjacent vehicles relative to the subject vehicle's path. By using the subject vehicle's expected course as a reference frame, the calculation remains manageable while enabling early detection of cutting-in intentions before vehicles actually enter the lane.
Data Source
AI summary
A driver assist apparatus includes: an information acquisition section that obtains a detected information of a different vehicle driving; a calculation section that calculates a risk estimation value representing a level of risk imposed on the subject vehicle by the different vehicle; a selection section that selects driver assist content corresponding to the risk estimation value; and a merging determination section that determines whether the different vehicle cuts into the expected course. Driver assist corresponding to the different vehicle cutting into the expected course is performed earlier than driver assist corresponding to a preceding vehicle driving on a lane on which the subject vehicle drives.