Parking Space Detection Using Search Lane Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automated parking systems are limited in their ability to accurately determine parking spaces, especially when insufficient sensor information is available, and often provide inaccurate results due to reliance on ultrasonic sensors or complex sensors like laser scanners, which are costly and require high computing power, and are typically restricted to pre-measured designated spaces.
Innovation Solution
A method that defines a search lane in front of the vehicle to align possible parking spaces, assigns measurement points to this lane, and determines parking spaces based on the alignment, allowing for efficient, robust, and flexible detection of parking spaces using non-local data, even with low-quality measurements, and minimizing computing power.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If ultrasonic sensors are used to measure parking spaces, then the system can detect parking spaces, but the measurement precision deteriorates when sensor information is insufficient
Solution Approach 1:
The patent divides the detection task into segments by using multiple sensor types (ultrasonic, optical, radar) to collect measurement points from different sources. Each sensor provides partial information, and the system integrates these segmented data sources to achieve complete and precise parking space detection, resolving the contradiction between detection capability and measurement precision.
Solution Approach 2:
The patent introduces an intermediary processing layer that collects measurement points from multiple sensors and uses geometric algorithms to determine parking space parameters. This intermediary layer transforms raw sensor data into accurate parking space measurements, enabling reliable detection while maintaining high precision even when individual sensor information is limited.
2Measurement precision
If complex sensors like laser scanners are used to detect parking spaces, then the measurement precision improves, but the device complexity and computing power requirements increase
Solution Approach 1:
The patent creates a universal detection system that can handle multiple sensor types (ultrasonic, optical, radar, camera) through a single integrated processing algorithm. The system processes measurement points from any sensor source using the same geometric evaluation methods, achieving high precision without requiring complex specialized hardware for each sensor type.
Solution Approach 2:
Instead of using expensive laser scanners directly, the patent creates a virtual model of the parking space by collecting measurement points from cheaper alternative sensors and reconstructing the parking space geometry through algorithms. This copying approach achieves comparable precision to laser scanners while using simpler, less expensive sensor equipment.
3Measurement precision
If the vehicle drives past the parking space to measure it, then the measurement precision improves, but the loss of time increases
Solution Approach 1:
The patent performs preliminary detection of parking spaces using sensors mounted on the vehicle while it approaches or is near the parking area. The system collects measurement points and evaluates potential parking spaces in advance, so when the vehicle needs to stop, the parking space is already identified and confirmed with sufficient precision, eliminating the need to drive past to measure.
Solution Approach 2:
The patent enables continuous parking space detection as the vehicle moves through the area, rather than requiring a separate measurement pass. The sensors continuously collect measurement points during normal vehicle operation, and the system continuously evaluates potential parking spaces, maintaining useful detection action throughout the vehicle's movement without requiring additional time for dedicated measurement.
4Device complexity
If GPS and pre-measured designated spaces are used, then the device complexity is reduced, but the adaptability deteriorates
Solution Approach 1:
The patent implements a dynamic detection system that adapts to different parking space configurations (parallel, perpendicular, diagonal, angled) based on the collected measurement points. The system evaluates the geometry of detected spaces and automatically adjusts its recognition criteria, enabling versatile detection across multiple parking space types while using a single flexible sensor system rather than dedicated systems for each type.
Solution Approach 2:
The patent changes the detection parameters and evaluation criteria based on the type of parking space being detected. The system adjusts geometric thresholds, orientation angles, and dimension requirements according to the detected configuration, enabling adaptability to various parking space types while maintaining a relatively simple sensor system that relies on flexible algorithmic parameter adjustment rather than hardware complexity.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
The present invention relates to a method for detecting a parking space from a number of measuring points, wherein each measuring point describes the position of an obstacle in the vicinity of a vehicle. This method comprises the steps of defining a search lane (4) along which possible parking spaces are arranged, determining an orientation (5) of possible parking spaces relative to the search lane (4), assigning the measuring points (2) to a corresponding point on the search lane taking into account the determined orientation (5) of possible parking spaces relative to the search lane (4), and identifying a parking space (1) based on the points of the search lane (4) to which a measuring point (2) has been assigned.