Parking Guidance Target Correction for Dividing Line Recognition Errors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing parking assistance systems face accuracy issues due to errors in recognizing dividing lines caused by changes in camera position and angle, which can lead to wheel cutting and reduced guidance precision, and adding sensors like a vehicle height sensor increases cost.
Innovation Solution
A parking assistance device that includes an image acquisition unit, a correction unit, and a guidance control unit to adjust the movement target position based on a reference position, correcting errors in dividing line recognition without additional physical configurations, and adjusting correction amounts based on distance and angle alignment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If dividing line recognition is performed using captured images, then parking space identification is enabled, but recognition accuracy deteriorates due to camera position and angle changes
Solution Approach 1:
The system performs preliminary action by setting a reference movement target position at the start of parking assistance when the vehicle is at a known initial position. This reference position is established before the camera position and angle may change during vehicle movement, allowing subsequent corrections to be made relative to this stable reference point rather than relying on continuously changing image-based recognition.
Solution Approach 2:
The system implements feedback by continuously correcting the movement target position based on the reference position throughout the parking maneuver. The correction unit adjusts the movement target position at predetermined control periods, using feedback from the vehicle's current position and the original reference position to compensate for errors introduced by camera position changes, thereby maintaining guidance accuracy.
2Measurement precision
If vehicle height sensor is added to improve recognition accuracy, then guidance precision is improved, but device cost increases
Solution Approach 1:
The system applies self-service by using its own existing imaging unit and control systems to perform the correction function. Rather than adding a vehicle height sensor, the system uses the captured images together with the reference movement target position and correction algorithms that are already part of the parking assistance system, making the existing components serve multiple functions including error compensation.
Solution Approach 2:
The system changes parameters by adjusting the movement target position based on calculated correction amounts rather than changing physical sensor configurations. The correction unit modifies the target position coordinates as the vehicle moves, dynamically adjusting the guidance parameters to compensate for recognition errors without requiring additional hardware sensors.
3Measurement precision
If movement target position is continuously corrected, then guidance accuracy is maintained, but control complexity increases
Solution Approach 1:
The correction amount is determined in advance based on the distance from the vehicle to the movement target position. The system calculates correction amounts at predetermined control periods before the vehicle reaches the target, using the reference position and current position to determine appropriate correction values, rather than making complex real-time adjustments during the entire parking maneuver.
Solution Approach 2:
The system applies dynamics by making the correction amount variable based on the vehicle's distance to the movement target position. As the vehicle approaches the target, the correction amount changes dynamically, allowing larger corrections when farther away and smaller corrections when closer, adapting the control strategy to the current state without requiring overly complex control logic.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A parking assistance device of an embodiment of the present invention includes, for example, an image acquisition unit (50) that acquires a captured image from an imaging unit (14) that images surrounding circumstances of a vehicle (10), a setting unit (58) that sets, at a start of parking assistance, a reference movement target position of the vehicle (10) in a parking space (42) defined by a pair of spaced dividing lines included in the captured image, and sets, after the start of the parking assistance, a movement target position of the vehicle (10) in the parking space (42) defined by the pair of dividing lines included in the captured image, at predetermined control periods, a correction unit (66) that corrects the movement target position, based on the reference movement target position, a position acquisition unit (54) that acquires the current position of the vehicle (10), and a guidance control unit (64) that performs control to guide the vehicle to the corrected movement target position, based on the corrected movement target position and the current position.