Automatic Parking System Using Smart Key Detection and Image Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automatic parking systems rely on drivers to interpret images from around-view monitoring systems, lacking assistance in determining optimal parking areas and requiring manual control, which can be challenging for inexperienced drivers and inefficient in narrow spaces.
Innovation Solution
An automatic parking system utilizing a smart key recognition processor, sensor unit, and controller to recognize parking areas through image recognition technology, calculate optimal parking paths, and control vehicle movements, including steering, acceleration, and braking, to facilitate remote and assisted parking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If a camera-based around view monitoring system is used to display images around the vehicle, then the driver can see the surrounding area, but the driver still has to manually determine parking conditions and control the vehicle
Solution Approach 1:
The system automatically detects parking spaces and calculates parking paths without requiring manual driver intervention. The control unit autonomously determines parking conditions from captured images, identifies suitable parking spaces, computes optimal parking paths, and controls the vehicle to execute the parking maneuver, making the system self-sufficient in performing the parking task
Solution Approach 2:
The patent replaces manual mechanical parking control with an automated image-processing and control system. The control unit uses image recognition algorithms to detect parking spaces and calculates parking paths, then automatically controls the vehicle's steering and movement, substituting the driver's manual operations with an automated computational control system
2Loss of information
If the driver manually determines parking conditions from AVM images, then the system provides visual information, but inexperienced drivers struggle to accurately identify optimal parking areas
Solution Approach 1:
The patent replaces the driver's manual visual assessment with an automated image recognition system. The control unit processes AVM images through algorithms that automatically detect parking spaces, calculate their dimensions, and determine optimality based on predefined criteria, eliminating the need for human interpretation and significantly improving detection accuracy and consistency
Solution Approach 2:
The system creates a digital representation of the parking environment by processing and analyzing AVM images. The control unit generates a computational model of the surrounding area, identifies parking spaces within this digital copy, and determines optimal parking locations based on the copied visual information, enabling accurate detection without relying on driver expertise
3Ease of operation
If automatic parking control is implemented, then parking ease is improved, but the system complexity increases with multiple sensors and control mechanisms
Solution Approach 1:
The control unit serves multiple functions within the system: it processes images from the AVM camera, detects parking spaces, calculates parking paths, and controls vehicle movement. By consolidating these diverse functions into a single multi-functional control unit, the system achieves automatic parking capability without proportionally increasing overall system complexity
Solution Approach 2:
The patent combines the image capture function, image processing function, path calculation function, and vehicle control function into an integrated automatic parking system. The control unit merges multiple operational tasks into a unified control mechanism, reducing the need for separate independent components and managing system complexity through functional integration
4Measurement precision
If the parking space recognition algorithm processes multiple consecutive images, then parking area detection accuracy is improved, but the processing time increases
Solution Approach 1:
The control unit continuously captures and pre-processes images from the AVM system even before a parking maneuver is initiated. By preparing image data in advance and maintaining a buffer of processed visual information, the system reduces the real-time processing burden when actual parking detection is needed, thereby improving detection accuracy without significantly increasing perceived processing time
Data Source
Figure 1
Figure 2
Figure 3
AI summary
There is provided an automatic parking system according to an exemplary embodiment of the present disclosure. The automatic parking system includes: a smart key recognition processor recognizing a smart key positioned in a first area; a sensor unit acquiring images around a subject vehicle if the smart key is recognized, and converting the acquired image into external images and synthesizing the images; a parking space recognizing sensor periodically receiving the external images and sequentially comparing the consecutive external images using an image recognition technology to recognize parking areas; and a controller calculating a moving path between a current position of the subject vehicle and an optimal parking area among the parking areas and controlling the subject vehicle based on the moving path, in which the parking space recognizing sensor may detect the parking areas in consideration of a length and a width of the subject vehicle.