Around-View Parking Recognition for Automated Space Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing automatic parking systems rely on drivers to interpret images from around-view monitoring systems to determine parking conditions, which can be cumbersome and require manual intervention.
Innovation Solution
An automatic parking system that utilizes a combination of camera processors, sensor processors, and a parking space recognizing unit to acquire images, measure distances, and recognize available parking areas, providing a recommended moving path to the driver and enabling remote operation for adjusting vehicle spacing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If a camera or ultrasonic sensor is used to detect parking areas, then parking automation is enabled, but the system complexity increases due to multiple sensors and image processing requirements
Solution Approach 1:
The system divides the complex parking detection task into separate functional modules: a camera processor for acquiring and synthesizing images from multiple angles, a sensor processor for measuring distances, and a parking space recognizing unit for identifying available parking areas. This segmentation allows each component to specialize in specific functions, reducing overall system complexity while maintaining high automation capability.
Solution Approach 2:
The patent introduces an around-view monitoring system as an intermediary that synthesizes images from multiple cameras into a comprehensive top-view representation of the parking environment. This intermediary processing layer simplifies the task of the parking space recognizing unit by providing pre-processed, easily interpretable visual data about available parking spaces, thereby reducing the complexity of the overall automation system.
2Measurement precision
If multiple imaging devices are installed to obtain images around the vehicle, then parking area recognition accuracy is improved, but the device complexity and cost increase
Solution Approach 1:
The patent combines images from multiple cameras positioned at different locations around the vehicle into a single synthesized around-view image that provides a comprehensive top-down perspective of the parking environment. This merging approach allows the system to achieve high recognition accuracy by utilizing data from multiple imaging devices while presenting the information in a unified, easily processable format that reduces computational complexity.
Solution Approach 2:
The system transforms two-dimensional images from multiple cameras into a three-dimensional virtual representation of the parking space by synthesizing around-view images that show the vehicle and surrounding environment from a top-down perspective. This dimensional transformation enables more accurate parking area recognition by providing spatial context that is not apparent in individual camera views, while the synthesized nature of the images keeps processing requirements manageable.
3Device complexity
If the driver manually determines parking conditions from AVM images, then system simplicity is maintained, but parking efficiency and convenience are reduced
Solution Approach 1:
The parking space recognizing unit automatically analyzes the synthesized around-view images and sensor data to identify available parking areas, calculate optimal parking positions, and determine safe maneuvering paths without requiring driver intervention. The system performs these functions autonomously by processing the visual and distance information it has already captured, thereby maintaining relative system simplicity while dramatically improving parking efficiency and convenience.
Solution Approach 2:
The system continuously monitors the parking environment using the around-view monitoring system and sensor processors, providing real-time feedback about detected parking spaces and potential obstacles to the controller. This feedback loop enables the automatic parking function to dynamically adjust the vehicle's parking maneuver based on actual environmental conditions, improving both efficiency and safety while keeping the overall system architecture relatively simple through reusable existing sensors.
4Speed
If automatic parking operation is performed without adjusting vehicle spacing, then parking speed is improved, but safety and convenience for vehicle exit are reduced
Solution Approach 1:
The controller dynamically adjusts the vehicle's parking position and spacing from surrounding vehicles during the automatic parking maneuver based on real-time sensor feedback. After the vehicle is parked, the system can automatically adjust the spacing to optimal values that balance quick parking with safety and convenience for driver exit, allowing the system to adapt to different parking scenarios without sacrificing speed.
Solution Approach 2:
The sensor processor pre-measures the distances to surrounding vehicles and the controller pre-calculates optimal spacing adjustments before finalizing the parking position. This preliminary action allows the system to plan the complete parking maneuver including spacing adjustments in advance, executing them efficiently in a single automated sequence that maintains high parking speed while ensuring safe and convenient final positioning.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An automatic parking system is provided. The automatic parking system includes a camera processor that acquires images around a subject vehicle, converts the acquired images into external images and synthesizes the external images. A sensor processor measured spaced distances between the subject vehicle and surrounding vehicles. A parking space recognizing unit periodically receives the spaced distances and the external images and comparing the consecutive external images with the spaced distances using an image recognition technology to recognize parking areas. A controller calculates a moving path between a current position of the subject vehicle and an optimal parking area and operates the subject vehicle based on the moving path.