Lane Change Detection Using Camera Angle Geometry
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing driver assistance systems for longitudinal guidance struggle to accurately detect vehicles cutting in or out of their lane, especially in close proximity, due to uncertainties in lane assignment and reliance on virtual driving paths rather than actual lane markings.
Innovation Solution
The method uses high-frequency two-dimensional camera images to determine the angle between virtual lines intersecting a detected vehicle and lane markings, allowing for precise detection of lateral movement and direction, without relying on the distance between vehicles, and normalizes this data to account for vehicle width, enabling improved detection of cutting-in or cutting-out processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If virtual driving paths are used for lane assignment, then the system can operate without direct lane marking detection, but detection accuracy deteriorates especially in close-range scenarios
Solution Approach 1:
The patent introduces lane markings as an intermediary reference element to improve lane assignment accuracy. By detecting lane markings in the camera image and using them as a reference, the system can more accurately determine the host vehicle's lane position and detect cut-in events, especially in close-range scenarios where virtual driving paths alone are insufficient.
2Length of stationary object
If radar or lidar sensors are used for object detection, then detection range is improved, but the ability to detect lateral movement and determine lane position deteriorates
Solution Approach 1:
The patent merges camera-based lane marking detection with radar/lidar-based object detection. The camera provides accurate lateral position information and lane assignment by detecting lane markings, while radar/lidar provides detection range and longitudinal distance. This combination allows the system to accurately detect both the presence and lateral position of surrounding vehicles.
3Reliability
If the system reacts late to cut-in vehicles, then false detections are reduced, but the usefulness of the detection deteriorates
Solution Approach 1:
The patent performs preliminary detection of lane markings and establishes the host vehicle's lane position in advance. By continuously monitoring the angular position of detected vehicles relative to the lane markings and calculating their lateral velocity, the system can identify cut-in events earlier and more accurately, providing timely warnings to the driver.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
The invention relates to a method for determining whether a vehicle ahead of the driver is merging or swerving, and to a corresponding vehicle. The driver's vehicle comprises a driver assistance system for longitudinal guidance and a camera that captures the area in front of the vehicle in the form of two-dimensional images. To determine whether a vehicle is merging or swerving, the angle (α) between two lines is determined in a recorded image. The first line runs from the camera (7) to the lane marking, and the second line runs from the camera (7) to the vehicle ahead.