Radar Parking Space Detection via Projection Profile Gaps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting parking spaces are not predictive and are prone to false detections due to various propagation effects, making it difficult to identify available spaces in advance and ensuring safe vehicle positioning.
Innovation Solution
A method using a radar device integrated in the vehicle to emit and receive signals, generating a distribution of radar signals in the X and Y coordinate directions, determining straight lines to define parking areas, and employing a projection profile and autoregressive prediction filter to predict gaps for identifying available parking spaces, which is robust against false detections.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If image-based methods are used to detect parking spaces, then the detection can be performed, but the detection is not predictive and only available when the vehicle is at the level of the parking spot
Solution Approach 1:
The radar sensor detects parking spaces in advance before the vehicle reaches them, performing the detection action preliminarily. The method projects radar signals laterally to identify available parking spaces at a distance, allowing the system to prepare for parking maneuvers before the vehicle is positioned directly over potential parking spots, thus resolving the contradiction between detection timing and predictive capability
2Loss of time
If beam sensor systems are used to illuminate the area in front of the vehicle, then potential parking spaces can be detected early, but the system requires sufficient processing time and safe deceleration distance
Solution Approach 1:
The patent replaces the mechanical beam sensor system with a radar-based detection system. Instead of using physical light beams and complex sensor arrays, the invention uses electromagnetic radar signals that can penetrate various conditions and provide detection without requiring the same level of system complexity or the same safety margins for deceleration
3Reliability
If radar signals are used to detect parking spaces, then detection can be performed, but false detections occur due to various propagation effects
Solution Approach 1:
The method employs feedback through cross-validation of radar signals with map data and geometric constraints. The system continuously refines its detection by comparing radar returns against expected parking space locations from map data and validating against geometric feasibility, thereby reducing false detections while maintaining reliable detection accuracy
Solution Approach 2:
The patent introduces map data and geometric constraint models as intermediary layers between the raw radar signals and the final parking space detection. These intermediaries filter and validate the radar information, eliminating false detections caused by propagation effects while preserving true parking space identifications
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 predictive and robust detection of parking spaces, enhancing driver assistance and automated parking capabilities by accurately identifying suitable parking spaces before reaching them, reducing the risk of false positives and negatives.
Implementation Method 1
a radar device which is integrated in the vehicle and comprises a transmitter with a signal generator and a receiver assembly and a receiver
Implementation Method 2
Radar signals reflected by parked vehicles and surrounding elements are received with the radar device
Data Source
AI summary
A method and a system (100) for detecting parking spaces (32) suitable for a vehicle (1). In order to determine a parking space (32), radar signals (11) are directed to a plurality of vehicles (301, 302, . . . 30M) parked in a parking area (13) and also to surrounding elements (34). The radar signals (12) reflected by the parked vehicles (301, 302, . . . 30M) and also the surrounding elements (34) are processed in a computing unit (15). A gap (29) in a calculated periodicity (51) of a projection profile (24) is determined with an autoregressive prediction filter (53). A prediction error function (26) has the highest value (55) at the location of a parking space (32).