Ultrasonic Object Detection Using Learning Model Feature Weighting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing object detection apparatuses using ultrasonic waves struggle to acquire highly accurate object information due to environmental factors affecting the reception signal, such as positional relationships and surrounding conditions.
Innovation Solution
The apparatus includes a reception unit for acquiring ultrasonic reception signals and an information processing unit that utilizes a learning model to extract temporal feature data from the signals. This model compresses the data, calculates feature quantities, and determines object shape by weighting these quantities with measurement information related to the waveform changes or object shape.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If only feature quantities from ultrasonic reception signals are used for object detection, then the device complexity is reduced, but the measurement precision of object shape information deteriorates due to environmental factors affecting the reception signal
Solution Approach 1:
The patent combines multiple types of measurement information (reception signal waveform features, distance information, and other correlated measurements) to determine object shape. By merging these different data sources and applying appropriate weighting, the system achieves high detection accuracy without requiring an overly complex device architecture.
Solution Approach 2:
The patent changes the parameter representation by extracting multiple feature quantities from the reception signal waveform (such as amplitude, duration, and shape characteristics) and combining them with other measurement parameters. This parameter transformation allows the system to compensate for environmental effects and improve measurement precision.
2Measurement precision
If multiple types of measurement information are combined to determine object shape, then the measurement precision improves, but the device complexity increases due to additional processing requirements
Solution Approach 1:
The patent performs preliminary processing of measurement information by extracting feature quantities from reception signals before combining them with other data. This advance preparation of data reduces the computational burden during the final object shape determination, thereby managing device complexity while maintaining high measurement precision.
Solution Approach 2:
The patent transforms raw measurement data into meaningful feature quantities through parameter changes, such as extracting waveform characteristics and normalizing data. This transformation simplifies the subsequent processing and combination of multiple information types, reducing overall system complexity.
3Loss of information
If feature quantities are extracted and compressed from reception signals, then the loss of information is reduced, but the processing time increases due to the complexity of feature extraction and calculation
Solution Approach 1:
The patent extracts only the essential feature quantities from the reception signals that are most relevant to object shape determination. By selectively extracting key features (such as amplitude, duration, and specific waveform characteristics) rather than processing the entire signal, the system preserves necessary information while reducing processing time.
Solution Approach 2:
The patent applies partial action by focusing on extracting and processing only the most critical feature quantities from the reception signal, rather than analyzing every aspect of the signal. This selective approach maintains sufficient information for accurate detection while minimizing processing time.
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 configuration enables the acquisition of highly accurate object information by considering both the feature quantities from the ultrasonic signals and the correlated measurement information, improving detection accuracy compared to systems relying solely on signal features.
Implementation Method 1
a reception signal corresponding to a reflected wave from the object of a transmission wave that is an ultrasonic wave
Data Source
AI summary
An object detection apparatus includes: a reception unit that acquires a reception signal corresponding to a reflected wave from the object of a transmission wave; and an information processing unit that acquires object information related to a shape of the object from a learning model by inputting a feature quantity and other measurement information of a waveform of the reception signal to the learned learning model in which machine learning to estimate the shape of an object has been performed. The learning model extracts, as temporal feature data, changes over time in a feature element in the reception signal based on the reception signal and a feature pattern of the object prescribed in advance, compresses the temporal feature data and acquires a plurality of feature quantities, and determines the shape of the object by calculating the plurality of feature quantities and the measurement information while weighting with a predetermined weight.