Ultrasonic-Camera Obstacle Verification for False Positive Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing obstacle detection systems using ultrasonic sensors and cameras often result in false positive detections, particularly with single-colored walls, leading to incorrect braking of vehicles, as current methods struggle to reliably verify obstacles using merged sensor data.
Innovation Solution
A method that involves an evaluation device receiving ultrasonic sensor data and camera images to determine free spaces by evaluating images with different criteria, verifying obstacles by comparing spatial boundaries and position with a threshold value, and using object recognition algorithms to accurately identify drivable areas and obstacles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If ultrasonic sensor data and camera images are merged to verify obstacles, then false positive detections are reduced, but the complexity of the detection system increases
Solution Approach 1:
The patent merges ultrasonic sensor data with camera image data to verify obstacle detections. The evaluation device combines results from both sensor types, where the ultrasonic sensor provides initial obstacle detection and the camera provides visual verification, reducing false positives through multi-modal sensor fusion
Solution Approach 2:
The evaluation device acts as an intermediary that processes and correlates data from both ultrasonic sensors and cameras. It mediates between the two sensor systems by comparing their results and making final obstacle verification decisions based on combined information
2Measurement precision
If object recognition algorithms are used to identify obstacles in camera images, then detection accuracy improves, but single-colored walls cannot be reliably detected
Solution Approach 1:
The patent substitutes object recognition algorithms with a free space analysis approach. Instead of relying on algorithms to identify and classify objects (which fail for single-colored walls), the system analyzes the spatial structure of free spaces in camera images and compares them with ultrasonic sensor data to verify obstacles
Solution Approach 2:
The patent transitions from object-based recognition (2D image analysis) to spatial dimension analysis (3D free space structure). By evaluating free spaces in multiple dimensions and comparing with ultrasonic ranging data, the system can detect obstacles that lack distinctive visual features
3Speed
If ultrasonic sensor data is used to detect obstacles, then detection speed is fast, but false positive detections occur
Solution Approach 1:
The ultrasonic sensor performs preliminary obstacle detection quickly to identify potential obstacles, and then the camera performs subsequent verification. This preliminary action allows fast initial detection while maintaining reliability through the follow-up verification step
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
This approach enhances the reliability of obstacle detection, preventing false positives and ensuring accurate identification of obstacles, thereby improving driver assistance systems by reducing incorrect braking and enhancing vehicle safety.
Implementation Method 1
at least one ultrasonic sensor of the motor vehicle... using the ultrasonic sensor data to identify the obstacle and a position of the obstacle
Implementation Method 2
at least one camera image from a camera of the motor vehicle... the obstacle detected using the ultrasonic sensor data is verified using the at least one camera image
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a method for detecting an obstacle (4) in an area (3) surrounding a motor vehicle (1), in which the following steps are performed by an evaluation unit (12) of the motor vehicle (1): - receiving ultrasonic sensor data from at least one ultrasonic sensor (6) of the motor vehicle (1) and from at least one camera image from at least one camera (7) of the motor vehicle (1), - detecting the obstacle (4) and determining a position of the obstacle (4) based on the ultrasonic sensor data, - verifying the obstacle (4) detected based on the ultrasonic sensor data using the at least one camera image, - determining a first clearance (18) by evaluating the at least one camera image with respect to a first criterion and determining a second clearance (19) by evaluating the at least one camera image with respect to a second criterion, wherein the clearances (18,19) an unobstructed sub-area (13) within the surrounding area (3) is described, -determining a first spatial boundary (23) of the first free space (18) and a second spatial boundary (24) of the second free space (19), -verifying the obstacle (4) if the first spatial boundary (23), the second spatial boundary (24), and the position differ from each other by no more than a predetermined threshold value. The invention also relates to an evaluation unit (12), a driver assistance system (2), and a motor vehicle (1).