Automated Driving Insertion Lane Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current driving assistance systems face difficulties in reliably detecting the presence of a vehicle on an insertion lane for roads with divided carriageways, especially when safety barriers are absent, which hinders safe and efficient automated driving.
Innovation Solution
A method that uses optical detection means to validate the presence of a vehicle on an insertion lane by analyzing demarcation line profiles, convergence, and adaptation of longitudinal and lateral control, adjusting speed and trajectory based on speed limits and adjacent traffic, while considering the absence of safety barriers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If radar analysis is used to detect discontinuity at safety barriers, then detection reliability is improved, but applicability deteriorates because most roads do not have such safety barriers
Solution Approach 1:
The patent introduces optical detection means (cameras) as an intermediary to detect demarcation lines and infer insertion lane presence. This mediator works universally on roads with painted demarcation lines, replacing the barrier-specific radar approach and enabling detection on roads without physical barriers.
Solution Approach 2:
The patent substitutes the mechanical/radar-based barrier detection system with an optical vision-based system. Instead of using radar to detect physical discontinuities at barriers, the system uses cameras to detect and analyze the geometry and convergence of painted demarcation lines, achieving universal applicability.
2Adaptability or versatility
If demarcation line analysis is used to detect insertion lane presence, then road type applicability is improved, but detection precision deteriorates due to uncertainty when safety barriers are absent
Solution Approach 1:
The patent transitions from analyzing only the lateral dimension of demarcation lines to analyzing their convergence in the longitudinal dimension. By extrapolating the profiles of both demarcation lines and checking if they converge at a specific distance, the system creates a new geometric criterion that significantly improves detection precision while maintaining universal applicability.
Solution Approach 2:
The system performs preliminary detection and validation steps before confirming insertion lane presence. It first detects demarcation lines, then analyzes their profiles, extrapolates their convergence, and only then validates the presence of an insertion lane. This multi-stage preliminary analysis reduces false positives and improves precision.
3Reliability
If automated driving control is adapted for insertion lane context, then driving safety is improved, but system complexity increases due to additional detection and control adaptations
Solution Approach 1:
The patent merges the insertion lane detection functionality with the existing automated driving control system. The same optical detection means used for general lane detection are also used for insertion lane detection, and the control adaptations are integrated into the existing longitudinal and lateral control modules, avoiding separate dedicated systems.
Solution Approach 2:
The automated driving control system is designed to handle multiple driving contexts universally. The same control module adapts its behavior based on the detected context (normal driving vs. insertion lane), adjusting longitudinal control for acceleration and lateral control for lane crossing authorization, without requiring separate dedicated control systems.
Data Source
Figure 1~2
Figure 3~4
AI summary
The invention concerns a method for detecting, during an automated driving phase of a vehicle, the presence of same in an access lane entering a road with separate carriageways, characterised in that it comprises the following steps: - checking, in front of said vehicle and over a predetermined anticipation distance, an upcoming convergence of the demarcation lines defining the traffic lane of said vehicle (200); - checking, when an upcoming convergence is detected and from the position of the vehicle and the map data, the existence of a road with separate carriageways in front of the vehicle and at a distance less than or equal to a maximum predetermined distance (300); and - validating, when the existence of such a road with separate carriageways is identified, the presence of said vehicle on such an access lane (600).