Parking Frame Correction Using Rear-View and AVM Coordinates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing automatic parking systems face errors due to discrepancies between the preset depth of a parking space and its actual depth, leading to inaccuracies in determining the target parking frame.
Innovation Solution
Utilizing an electronic rear-view mirror and an around view monitor (AVM) to capture and process images, the method determines a target parking frame by adjusting far end coordinates based on real-time images from the rear-view mirror and combining them with near end points from the AVM images, thereby correcting the far end point of the reference parking frame.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If the AVM constructs the target parking frame based on a preset depth of the parking space, then the parking frame can be determined quickly, but the error between the target parking frame and the actual parking frame increases
Solution Approach 1:
The system performs preliminary action by capturing images of the parking space before the vehicle enters, using the electronic rear-view mirror to obtain the far end coordinates and the AVM to obtain the near end coordinates. This preliminary image capture and coordinate extraction allows the system to pre-calculate the actual parking frame dimensions, avoiding the need to use preset depth values during the actual parking maneuver, thus resolving the contradiction between quick determination and high accuracy.
Solution Approach 2:
The system implements feedback by using the electronic rear-view mirror to capture real-time images of the parking space and comparing the captured far end coordinates with the near end coordinates from the AVM. This feedback mechanism allows the system to continuously adjust and refine the parking frame determination based on actual visual information, ensuring high accuracy while maintaining efficient processing through iterative refinement.
2Device complexity
If the AVM uses a preset fixed depth value to determine the parking frame, then the system complexity is reduced, but the reliability of the parking frame determination decreases
Solution Approach 1:
The system applies universality by making the electronic rear-view mirror serve multiple functions: it acts as both a driver assistance device for monitoring blind spots and as a measurement device for capturing parking space images and determining far end coordinates. This multi-functionality allows the system to achieve reliable parking frame determination without adding dedicated hardware, thus maintaining low system complexity while improving reliability.
Solution Approach 2:
The system implements self-service by using the vehicle's existing electronic rear-view mirror to capture images and provide measurement data for parking frame determination. The rear-view mirror system serves itself by utilizing its imaging capability for dual purposes: driver assistance and parking guidance. This self-service approach eliminates the need for separate dedicated sensors or cameras, reducing system complexity while ensuring reliable data acquisition.
3Area of stationary object
If the AVM only captures partial images of the parking space, then the imaging coverage requirement is reduced, but the measurement precision of the parking frame decreases
Solution Approach 1:
The system applies dimensionality change by utilizing the electronic rear-view mirror to capture the far end of the parking space in a different spatial dimension (side/rear view) compared to the AVM's front view. This multi-dimensional approach allows the system to obtain complete coordinate information (both near end and far end) without requiring any single camera to capture the entire parking space, thus resolving the contradiction between limited imaging coverage and high measurement precision.
Data Source
Figure 1~2
Figure 3~5
Figure 6~7
AI summary
A method for determining a parking frame includes: determining first far end coordinates based on a first image captured by the electronic rear-view mirror (11), the first far end coordinates being coordinates of a target far end point of the parking space in the first image; obtaining second far end coordinates of the target far end point in a reference parking frame, the reference parking frame being generated based on a second image obtained by the AVM (12), and the second image including two near end points of the parking space; adjusting the second far end coordinates according to the first far end coordinates to obtain third far end coordinates; and determining a target parking frame of the parking space based on the third far end coordinates and coordinates of the two near end points in the reference parking frame.