Ultrasonic Sensor Data Processing with Neural Network
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing ultrasonic sensor data processing methods struggle to accurately determine the position and height of objects in real-time, particularly in complex environments, limiting their effectiveness in applications like obstacle detection and vehicle navigation.
Innovation Solution
A method utilizing a neural network, specifically a U-Net type convolutional neural network, processes ultrasonic signal data to generate a two-dimensional data set representing potential object positions and heights by analyzing echo curves and signal features, enhancing spatial resolution and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional ultrasonic sensor data processing methods are used, then the system is simple and easy to implement, but the position and height detection precision is insufficient
Solution Approach 1:
The patent introduces a neural network as an intermediary component between the ultrasonic sensor and the detection system. The neural network processes raw ultrasonic signals and transforms them into enhanced position and height data, enabling precise detection without requiring complex traditional signal processing algorithms. This intermediary approach improves measurement precision while keeping the overall system architecture relatively simple.
2Speed
If real-time processing is implemented, then the response time is reduced, but the processing accuracy in complex environments deteriorates
Solution Approach 1:
The patent employs a neural network that processes multiple parameters from ultrasonic signals simultaneously, including echo amplitude, time of flight, and frequency information. By changing the parameter representation and processing approach, the system achieves real-time computation while maintaining high accuracy in complex environments through the neural network's ability to weigh different parameters optimally.
3Measurement precision
If the spatial resolution is increased, then the detection accuracy is improved, but the data processing time increases
Solution Approach 1:
The patent replaces traditional mechanical or algorithmic signal processing with a neural network-based approach. The neural network efficiently processes spatial data and generates high-resolution maps of object positions and heights without requiring computationally intensive traditional methods. This substitution enables high spatial resolution processing in real-time, significantly reducing processing time compared to conventional approaches.
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
The method provides precise object detection and classification, enabling improved obstacle avoidance and navigation by accurately mapping object presence and height, thereby enhancing safety and functionality in vehicles.
Implementation Method 1
a received portion of an ultrasonic signal that has been transmitted by a transmitter and that has been reflected and/or scattered by at least one object before it is received by a receiver
Implementation Method 2
processing the data set using a, for example artificial, neural network
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
A method for processing data associated with at least one ultrasonic sensor, comprising: receiving at least one signal characterizing at least a portion of a transmitted ultrasonic signal, determining, based on the at least one signal, a two-dimensional data set having a plurality of cells characterizing potential positions of an object relative to an ultrasonic sensor associated with the at least one signal, processing the data set using a neural network.