Ultrasonic Sensor Object Classification via Neural Network Feature Vectors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
State-of-the-art ultrasonic sensors in automotive and industrial applications primarily focus on distance measurement using transit time, underutilizing the time signal for object classification, leading to impaired accuracy in differentiating objects and determining their dimensions due to limited data processing and transmission.
Innovation Solution
A method involving the transmission and processing of ultrasonic signals to extract relevant time-limited segments, transforming them into two-dimensional feature vectors, and using a neural network for object classification, incorporating distance information and advanced signal processing techniques like filtering and convolutional layers to enhance classification accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If only amplitude or correlation values above threshold are transmitted, then data transmission volume is reduced, but object classification accuracy deteriorates
Solution Approach 1:
The patent extracts only the relevant time-limited signal portion containing the echo information, transforming it into a compact two-dimensional feature vector. This extraction method retains the essential classification information while reducing data volume, resolving the contradiction between information loss and classification accuracy.
Solution Approach 2:
The patent transforms the one-dimensional time signal into a two-dimensional feature vector representation. This dimensional transformation preserves the temporal and amplitude characteristics of the echo signal in a compact format that maintains classification accuracy while reducing data transmission requirements.
2Measurement precision
If the entire time signal is processed and transmitted, then object classification accuracy is improved, but data transmission volume and processing complexity increase
Solution Approach 1:
The patent extracts only the relevant time-limited signal portion containing the echo information, transforming it into a compact two-dimensional feature vector. This extraction method retains the essential classification information while reducing data volume, resolving the contradiction between information loss and classification accuracy.
Solution Approach 2:
The patent segments the entire time signal into distinct portions, selecting only the relevant time-limited segment that contains the echo information. This segmentation approach processes only the necessary data portion, reducing computational complexity while maintaining classification accuracy.
3Device complexity
If traditional threshold-based echo detection is used, then system complexity is reduced, but object differentiation capability deteriorates
Solution Approach 1:
The patent transforms the one-dimensional time signal into a two-dimensional feature vector representation. This dimensional transformation preserves the temporal and amplitude characteristics of the echo signal in a compact format that maintains classification accuracy while reducing data transmission requirements.
Solution Approach 2:
The patent creates a multi-functional processing framework where the same signal processing pipeline serves both simple distance measurement and complex object classification tasks. The two-dimensional feature vector representation provides universal information that supports multiple application scenarios from basic detection to advanced classification.
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 enables efficient and reliable classification of objects, improving the performance of ultrasonic sensors by determining object classes, such as static vs. moving objects or drivable vs. non-drivable surfaces, enhancing safety and automation in driving functions.
Implementation Method 1
Ultrasonic sensors operate according to the well-known pulse-echo principle. An electrical signal excites a transducer to vibrations on its membrane, which are emitted as sound. The sound travels through the air until it meets an object in the surroundings. The surface of the object reflects the sound, which causes backscattering in the direction of the ultrasonic sensor.
Implementation Method 2
The sound travels through the air until it meets an object in the surroundings. The known speed of sound propagation can be used to ascertain the distance to the backscattering object.
Implementation Method 3
The surface of the object reflects the sound, which causes backscattering in the direction of the ultrasonic sensor.
Implementation Method 4
When the backscattered sound hits the membrane, it causes it to vibrate and creates an electrical signal at the piezo element.
Data Source
AI summary
A method for classifying an object to be detected with at least one ultrasonic sensor. The method includes: transmitting a first signal using the ultrasonic sensor to the object; receiving a second signal using the ultrasonic sensor, wherein the second signal is a backscattered signal from the object; processing the second signal into a digital signal; extracting a selected signal portion from the digital signal, the selected signal portion representing a relevant and time-limited time segment from the digital signal; transforming the selected signal portion into a two-dimensional feature vector; feeding the two-dimensional feature vector into a neural network as at least one input variable; determining object class information for the object using the neural network, wherein, based on the at least one input variable, the neural network produces an output variable which indicates a probability value for at least one defined object class for the one object.