3D Parking Space Detection Using Foresighted Sensor Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle parking systems require vehicles to drive past potential parking spaces to assess suitability, wasting time and relying on incomplete data, as they cannot accurately detect and classify parking spaces before arrival.
Innovation Solution
A method and device that use foresighted three-dimensional data recording with sensors providing horizontal and vertical depth information to classify objects and determine the geometry of potential parking spaces before arrival, allowing early detection and classification of suitable spaces, including those partially or fully delimited by lines or objects, and enabling automatic or assisted parking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If the vehicle drives past the potential parking space to assess suitability, then the detection can be performed with simple sensors, but time is wasted and the detection must rely on incomplete data
Solution Approach 1:
The system performs preliminary detection and classification of parking spaces before the vehicle arrives. Sensors mounted on the vehicle capture images and depth information of potential parking spaces in advance, allowing the evaluation unit to assess suitability before the driver needs to make a decision, thus eliminating the need to drive past spaces to gather information
2Loss of information
If three-dimensional data recording is performed foresightedly, then complete information about parking spaces is available before arrival, but the device complexity increases
Solution Approach 1:
The system uses a camera that can serve multiple functions: capturing two-dimensional images for visual recognition and providing depth information through techniques like stereoscopy or structured light. The evaluation unit performs multiple tasks including object detection, classification, and parking space evaluation using the same sensor data, reducing the need for separate specialized sensors for each function
Solution Approach 2:
The system introduces an evaluation unit that acts as an intermediary between the sensors and the driver. This unit processes complex sensor data, classifies objects, determines parking space geometry, and presents simplified results to the driver, reducing the perceived complexity while enabling comprehensive three-dimensional analysis
3Reliability
If all surrounding areas are recorded to ensure complete detection, then no parking space is missed, but unnecessary background information increases processing load
Solution Approach 1:
The evaluation unit extracts only the relevant information needed for parking space detection from the complete sensor data. It identifies and focuses on objects that could delimit parking spaces (curbs, walls, other vehicles) while filtering out irrelevant background elements, thus maintaining reliable detection without processing unnecessary data
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables early detection and classification of parking spaces, allowing for time-saving and efficient parking by providing complete information before arrival, supporting the driver and enabling partially or fully automatic parking processes, even in complex scenarios.
Implementation Method 1
a sensor which can move with the vehicle for recording three-dimensional data of an area surrounding the vehicle... a sensor that supplies horizontally and vertically resolved depth information
Implementation Method 2
Such three-dimensional data can be recorded in particular with a sensor that supplies horizontally and vertically resolved depth information
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
The method involves recording and collecting three dimensional data for an environment of motor vehicles (1, 2, 3). Objects surrounding the vehicles are classified into a parking area limited object and a non parking limited object. A spatial area adapted as a parking area for the vehicles is determined in the environment of the vehicles. Data of the spatial area adapted as the parking area is outputted to a driver or a superordinate system. The data outputted over the spatial area adapted as the parking area is obtained from an area in front of a driver of the vehicle at the spatial area. An independent claim is also included for a device for detecting and/or measuring a parking space, comprising a sensor.